Merge branch 'master' into latvian

.gitignore

@@ -66,11 +66,13 @@ depcomp
 install-sh
 libtool
 ltmain.sh
-m4/
 missing
 stamp-h1
 .dirstamp
 
+m4/*
+!m4/ax_check_compile_flag.m4
+
 # libraries
 
 docs/speak_lib.h
@@ -96,6 +98,7 @@ src/speak-ng
 
 tests/*.test
 !tests/languages.test
+!tests/phoneme-output.test
 
 espeak-ng.pc
 

CHANGELOG.md

@@ -7,7 +7,21 @@
 
 The espeak-ng project is a fork of the espeak project.
 
-### 1.49.2 - (In Development)
+### 1.49.3 - (In Development)
+
+bug fixes:
+
+*  Fix several issues with the emoji support.
+
+updated languages:
+
+*  af (Afrikaans) -- Christo de Klerk
+*  fi (Finnish) -- Juho Hiltunen
+*  it (Italian) -- chrislm
+*  lv (Latvian) -- Valdis Vitolins
+*  ur (Urdu) -- Ejaz Shah
+
+### 1.49.2 - 2017-09-24
 
 *  Provide more validation when reading phoneme tables and voice/language files.
 *  Removed support for phoneme equivalence tables. These were disabled in the
@@ -25,6 +39,7 @@ bug fixes:
 *  Fix running `make clean ; make`.
 *  Fix reading stdin buffers larger than 1000.
 *  Fixed various language and parent BCP 47 codes (`jp` is now `ja`).
+*  Fixed several crashes and bugs in `espeak_SetVoiceByName/Properties`.
 
 new languages:
 
@@ -32,7 +47,7 @@ new languages:
 *  fr-CH (French (Switzerland)) -- Claude Beazley
 *  kok (Konkani) -- Vardhan
 *  mi (Māori) -- Graham Oliver
-*  sd (Sindhi, Arabic script) -- Vardhan
+*  sd (Sindhi, Arabic script) -- Vardhan, Ejaz Shah
 
 updated languages:
 
@@ -90,10 +105,14 @@ Unicode Emoji support (from the Unicode Emoji 5.0, and CLDR 31.0.1 data files):
 *  lt (Lithuanian)
 *  lv (Latvian)
 *  mk (Macedonian)
+*  ml (Malayalam)
 *  mr (Marathi)
 *  ms (Malay)
+*  my (Myanmar/Burmese)
+*  ne (Nepali)
 *  nl (Dutch)
 *  or (Oriya)
+*  pa (Punjabi)
 *  pl (Polish)
 *  pt (Portuguese) -- Brazil
 *  ro (Romanian)
@@ -105,6 +124,7 @@ Unicode Emoji support (from the Unicode Emoji 5.0, and CLDR 31.0.1 data files):
 *  sr (Serbian)
 *  sv (Swedish)
 *  sw (Swahili)
+*  ta (Tamil)
 *  te (Telugu)
 *  tr (Turkish)
 *  ur (Urdu)

Makefile.am

@@ -131,8 +131,7 @@ espeak_ng_includedir = $(includedir)/espeak-ng
 espeak_ng_include_HEADERS = \
 	src/include/espeak-ng/encoding.h \
 	src/include/espeak-ng/espeak_ng.h \
-	src/include/espeak-ng/speak_lib.h \
-	src/include/espeak-ng/tokenizer.h
+	src/include/espeak-ng/speak_lib.h
 
 lib_LTLIBRARIES += src/libespeak-ng.la
 
@@ -171,7 +170,6 @@ src_libespeak_ng_la_SOURCES = \
 	src/libespeak-ng/synthdata.c \
 	src/libespeak-ng/synthesize.c \
 	src/libespeak-ng/synth_mbrola.c \
-	src/libespeak-ng/tokenizer.c \
 	src/libespeak-ng/translate.c \
 	src/libespeak-ng/tr_languages.c \
 	src/libespeak-ng/voices.c \
@@ -233,12 +231,6 @@ noinst_bin_PROGRAMS += tests/encoding.test
 tests_encoding_test_LDADD   = src/libespeak-ng.la
 tests_encoding_test_SOURCES = tests/encoding.c
 
-noinst_bin_PROGRAMS += tests/tokenizer.test
-
-tests_tokenizer_test_CFLAGS  = -Isrc/libespeak-ng ${AM_CFLAGS}
-tests_tokenizer_test_LDADD   = src/libespeak-ng.la
-tests_tokenizer_test_SOURCES = tests/tokenizer.c
-
 noinst_bin_PROGRAMS += tests/readclause.test
 
 tests_readclause_test_CFLAGS  = -Isrc/libespeak-ng ${AM_CFLAGS}
@@ -256,9 +248,9 @@ tests_api_test_SOURCES = tests/api.c
 	@ESPEAK_DATA_PATH=$(PWD) $< && echo "  PASSED    $<"
 
 check:	tests/encoding.check \
-	tests/tokenizer.check \
 	tests/readclause.check \
 	tests/api.check \
+	tests/phoneme-output.check \
 	tests/languages.check
 
 ##### phoneme data:
@@ -398,6 +390,7 @@ android/res/raw/espeakdata.zip: espeak-ng-data/phontab dictionaries
 	rm -f $@
 	find espeak-ng-data/{intonations,phondata,phonindex,phontab} | zip -@ $@
 	find espeak-ng-data/*_dict | zip -@ $@
+	find espeak-ng-data/lang -type f | zip -@ $@
 	find espeak-ng-data/voices -type f | grep -vF "/mb/" | zip -@ $@
 
 android/res/raw/espeakdata_version: android/res/raw/espeakdata.zip
@@ -672,7 +665,7 @@ mk: espeak-ng-data/mk_dict
 espeak-ng-data/mk_dict: dictsource/mk_list dictsource/mk_rules dictsource/mk_extra dictsource/mk_emoji
 
 ml: espeak-ng-data/ml_dict
-espeak-ng-data/ml_dict: dictsource/ml_list dictsource/ml_rules dictsource/ml_extra
+espeak-ng-data/ml_dict: dictsource/ml_list dictsource/ml_rules dictsource/ml_extra dictsource/ml_emoji
 
 mr: espeak-ng-data/mr_dict
 espeak-ng-data/mr_dict: dictsource/mr_list dictsource/mr_rules dictsource/mr_extra dictsource/mr_emoji
@@ -684,13 +677,13 @@ mt: espeak-ng-data/mt_dict
 espeak-ng-data/mt_dict: dictsource/mt_list dictsource/mt_rules dictsource/mt_extra
 
 my: espeak-ng-data/my_dict
-espeak-ng-data/my_dict: dictsource/my_list dictsource/my_rules dictsource/my_extra
+espeak-ng-data/my_dict: dictsource/my_list dictsource/my_rules dictsource/my_extra dictsource/my_emoji
 
 nci: espeak-ng-data/nci_dict
 espeak-ng-data/nci_dict: dictsource/nci_list dictsource/nci_rules dictsource/nci_extra
 
 ne: espeak-ng-data/ne_dict
-espeak-ng-data/ne_dict: dictsource/ne_list dictsource/ne_rules dictsource/ne_extra
+espeak-ng-data/ne_dict: dictsource/ne_list dictsource/ne_rules dictsource/ne_extra dictsource/ne_emoji
 
 nl: espeak-ng-data/nl_dict
 espeak-ng-data/nl_dict: dictsource/nl_list dictsource/nl_rules dictsource/nl_extra dictsource/nl_emoji
@@ -705,7 +698,7 @@ or: espeak-ng-data/or_dict
 espeak-ng-data/or_dict: dictsource/or_list dictsource/or_rules dictsource/or_extra dictsource/or_emoji
 
 pa: espeak-ng-data/pa_dict
-espeak-ng-data/pa_dict: dictsource/pa_list dictsource/pa_rules dictsource/pa_extra
+espeak-ng-data/pa_dict: dictsource/pa_list dictsource/pa_rules dictsource/pa_extra dictsource/pa_emoji
 
 pap: espeak-ng-data/pap_dict
 espeak-ng-data/pap_dict: dictsource/pap_list dictsource/pap_rules dictsource/pap_extra
@@ -761,7 +754,7 @@ sw: espeak-ng-data/sw_dict
 espeak-ng-data/sw_dict: dictsource/sw_list dictsource/sw_rules dictsource/sw_extra dictsource/sw_emoji
 
 ta: espeak-ng-data/ta_dict
-espeak-ng-data/ta_dict: dictsource/ta_list dictsource/ta_rules dictsource/ta_extra
+espeak-ng-data/ta_dict: dictsource/ta_list dictsource/ta_rules dictsource/ta_extra dictsource/ta_emoji
 
 te: espeak-ng-data/te_dict
 espeak-ng-data/te_dict: dictsource/te_list dictsource/te_rules dictsource/te_extra dictsource/te_emoji

README.md

@@ -49,11 +49,11 @@ The following platforms are supported:
 
 ### Binaries
 
-The Windows version of eSpeak NG 1.49.1 is available as:
+The Windows version of eSpeak NG 1.49.2 is available as:
 
-*  [espeak-ng-x64.msi](https://github.com/espeak-ng/espeak-ng/releases/download/1.49.1/espeak-ng-x64.msi) --
+*  [espeak-ng-x64.msi](https://github.com/espeak-ng/espeak-ng/releases/download/1.49.2/espeak-ng-x64.msi) --
    64-bit Windows installer
-*  [espeak-ng-x86.msi](https://github.com/espeak-ng/espeak-ng/releases/download/1.49.1/espeak-ng-x86.msi) --
+*  [espeak-ng-x86.msi](https://github.com/espeak-ng/espeak-ng/releases/download/1.49.2/espeak-ng-x86.msi) --
    32-bit Windows installer
 
 You also need to install the [Visual C++ Redistributable for Visual Studio 2015](https://www.microsoft.com/en-us/download/details.aspx?id=48145).
@@ -69,7 +69,7 @@ To build eSpeak NG on Windows, you will need:
 1. a copy of Visual Studio 2013 or later, such as the Community Edition;
 2. the Windows 8.1 SDK;
 3. the [WiX installer](http://wixtoolset.org) plugin;
-4. the [pcaudiolib](http://github.com/rhdunn/pcaudiolib) project checked out to
+4. the [pcaudiolib](https://github.com/espeak-ng/pcaudiolib) project checked out to
    `src` (as `src/pcaudiolib`).
 
 You can then open and build the `src/windows/espeak-ng.sln` solution in Visual
@@ -104,9 +104,9 @@ In order to build eSpeak NG, you need:
 
 Optionally, you need:
 
-1.  the [pcaudiolib](https://github.com/rhdunn/pcaudiolib) development library
+1.  the [pcaudiolib](https://github.com/espeak-ng/pcaudiolib) development library
     to enable audio output;
-3.  the [sonic](https://github.com/waywardgeek/sonic) development library to
+3.  the [sonic](https://github.com/espeak-ng/sonic) development library to
     enable sonic audio speed up support;
 4.  the `ronn` man-page markdown processor to build the man pages.
 
@@ -391,8 +391,8 @@ The `ieee80.c` implementation is taken directly from
 [ToFromIEEE.c.txt](http://www.realitypixels.com/turk/opensource/ToFromIEEE.c.txt)
 which has been made available for use in Open Source applications per the
 [license statement](COPYING.IEEE) on http://www.realitypixels.com/turk/opensource/.
-The only modification made to the code is to comment out the `TEST_FP` define
-to make it useable in the eSpeak NG library.
+The only modifications made to the code is to comment out the `TEST_FP` define
+to make it useable in the eSpeak NG library, and to fix compiler warnings.
 
 The `getopt.c` compatibility implementation for getopt support on Windows is
 taken from the NetBSD `getopt_long` implementation, which is licensed under a

android/build.gradle

@@ -3,7 +3,7 @@ buildscript {
         mavenCentral()
     }
     dependencies {
-        classpath 'com.android.tools.build:gradle:1.2.3'
+        classpath 'com.android.tools.build:gradle:2.3.3'
     }
 }
 
@@ -18,12 +18,12 @@ dependencies {
 }
 
 android {
-    buildToolsVersion '19.1'
-    compileSdkVersion 23
+    buildToolsVersion '25.0.3'
+    compileSdkVersion 26
 
     defaultConfig {
         minSdkVersion 14
-        targetSdkVersion 23
+        targetSdkVersion 26
         versionCode 20
         versionName "1.48.15"
     }
@@ -33,6 +33,7 @@ android {
             manifest.srcFile 'AndroidManifest.xml'
             java.srcDirs = ['src']
             res.srcDirs = ['res']
+            jniLibs.srcDir 'libs'
         }
         androidTest.setRoot('eSpeakTests')
         androidTest {
@@ -44,8 +45,3 @@ android {
         abortOnError false
     }
 }
-
-tasks.withType(com.android.build.gradle.tasks.PackageApplication) {
-    pkgTask -> pkgTask.jniFolders = new HashSet<File>()
-    pkgTask.jniFolders.add(new File(projectDir, 'libs'))
-}

android/eSpeakTests/src/com/reecedunn/espeak/test/SpeechSynthesisTest.java

@@ -260,7 +260,7 @@ public class SpeechSynthesisTest extends TextToSpeechTestCase
 
     public void testMatchVoiceWithLanguageAndCountry()
     {
-        final Voice voice = getVoice("fr-fr"); // language="fr" country="fr" variant=""
+        final Voice voice = getVoice("fr-be"); // language="fr" country="BE" variant=""
         assertThat(voice, is(notNullValue()));
 
         assertThat(voice.match(de), isTtsLangCode(TextToSpeech.LANG_NOT_SUPPORTED));
@@ -269,23 +269,23 @@ public class SpeechSynthesisTest extends TextToSpeechTestCase
         assertThat(voice.match(de_CH_1901), isTtsLangCode(TextToSpeech.LANG_NOT_SUPPORTED));
 
         assertThat(voice.match(fr), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
-        assertThat(voice.match(fr_FR), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
-        assertThat(voice.match(fr_BE), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        assertThat(voice.match(fr_FR), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        assertThat(voice.match(fr_BE), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
         assertThat(voice.match(fr_1694acad), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
-        assertThat(voice.match(fr_FR_1694acad), isTtsLangCode(TextToSpeech.LANG_COUNTRY_AVAILABLE));
-        assertThat(voice.match(fr_BE_1694acad), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        assertThat(voice.match(fr_FR_1694acad), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        assertThat(voice.match(fr_BE_1694acad), isTtsLangCode(TextToSpeech.LANG_COUNTRY_AVAILABLE));
 
         assertThat(voice.match(fra), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
-        assertThat(voice.match(fra_FRA), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
-        assertThat(voice.match(fra_BEL), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        assertThat(voice.match(fra_FRA), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        assertThat(voice.match(fra_BEL), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
         assertThat(voice.match(fra_1694acad), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
-        assertThat(voice.match(fra_FRA_1694acad), isTtsLangCode(TextToSpeech.LANG_COUNTRY_AVAILABLE));
-        assertThat(voice.match(fra_BEL_1694acad), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        assertThat(voice.match(fra_FRA_1694acad), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        assertThat(voice.match(fra_BEL_1694acad), isTtsLangCode(TextToSpeech.LANG_COUNTRY_AVAILABLE));
     }
 
     public void testMatchVoiceWithLanguageCountryAndVariant()
     {
-        final Voice voice = getVoice("en-sc"); // language="en" country="GB" variant="scotland"
+        final Voice voice = getVoice("en-gb-scotland"); // language="en" country="GB" variant="scotland"
         assertThat(voice, is(notNullValue()));
 
         assertThat(voice.match(de), isTtsLangCode(TextToSpeech.LANG_NOT_SUPPORTED));

android/eSpeakTests/src/com/reecedunn/espeak/test/TextToSpeechServiceTest.java

@@ -99,10 +99,10 @@ public class TextToSpeechServiceTest extends AndroidTestCase
     }
 
     public void testOnLoadLanguage() {
-        assertThat(mService.onLoadLanguage("eng", "", ""), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
-        checkLanguage(mService.onGetLanguage(), "eng", "", "");
+        assertThat(mService.onLoadLanguage("eng", "", ""), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        checkLanguage(mService.onGetLanguage(), "eng", "GBR", "");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("en"));
+        assertThat(mService.getActiveVoice().name, is("en-gb"));
 
         assertThat(mService.onLoadLanguage("eng", "USA", ""), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
         checkLanguage(mService.onGetLanguage(), "eng", "USA", "");
@@ -112,120 +112,120 @@ public class TextToSpeechServiceTest extends AndroidTestCase
         assertThat(mService.onLoadLanguage("eng", "GBR", "scotland"), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
         checkLanguage(mService.onGetLanguage(), "eng", "GBR", "scotland");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("en-sc"));
+        assertThat(mService.getActiveVoice().name, is("en-gb-scotland"));
 
         assertThat(mService.onLoadLanguage("eng", "USA", "rp"), isTtsLangCode(TextToSpeech.LANG_COUNTRY_AVAILABLE));
         checkLanguage(mService.onGetLanguage(), "eng", "USA", "");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
         assertThat(mService.getActiveVoice().name, is("en-us"));
 
-        assertThat(mService.onLoadLanguage("eng", "", "scotland"), isTtsLangCode(TextToSpeech.LANG_COUNTRY_AVAILABLE));
-        checkLanguage(mService.onGetLanguage(), "eng", "", "");
+        assertThat(mService.onLoadLanguage("eng", "", "scotland"), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        checkLanguage(mService.onGetLanguage(), "eng", "GBR", "");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("en"));
+        assertThat(mService.getActiveVoice().name, is("en-gb"));
 
         assertThat(mService.onLoadLanguage("eng", "FRA", "rp"), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
-        checkLanguage(mService.onGetLanguage(), "eng", "", "");
+        checkLanguage(mService.onGetLanguage(), "eng", "GBR", "");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("en"));
+        assertThat(mService.getActiveVoice().name, is("en-gb"));
 
         assertThat(mService.onLoadLanguage("eng", "FRA", ""), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
-        checkLanguage(mService.onGetLanguage(), "eng", "", "");
+        checkLanguage(mService.onGetLanguage(), "eng", "GBR", "");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("en"));
+        assertThat(mService.getActiveVoice().name, is("en-gb"));
 
         assertThat(mService.onLoadLanguage("ine", "", ""), isTtsLangCode(TextToSpeech.LANG_NOT_SUPPORTED));
-        checkLanguage(mService.onGetLanguage(), "eng", "", "");
+        checkLanguage(mService.onGetLanguage(), "eng", "GBR", "");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("en"));
+        assertThat(mService.getActiveVoice().name, is("en-gb"));
     }
 
     public void testOnIsLanguageAvailable() {
-        assertThat(mService.onLoadLanguage("vie", "VNM", "saigon"), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
+        assertThat(mService.onLoadLanguage("vie", "VNM", "central"), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
 
-        assertThat(mService.onIsLanguageAvailable("eng", "", ""), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        assertThat(mService.onIsLanguageAvailable("eng", "", ""), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
         assertThat(mService.onIsLanguageAvailable("eng", "USA", ""), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
         assertThat(mService.onIsLanguageAvailable("eng", "GBR", "scotland"), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
         assertThat(mService.onIsLanguageAvailable("eng", "USA", "rp"), isTtsLangCode(TextToSpeech.LANG_COUNTRY_AVAILABLE));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
-        assertThat(mService.onIsLanguageAvailable("eng", "", "scotland"), isTtsLangCode(TextToSpeech.LANG_COUNTRY_AVAILABLE));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        assertThat(mService.onIsLanguageAvailable("eng", "", "scotland"), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
         assertThat(mService.onIsLanguageAvailable("eng", "FRA", "rp"), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
         assertThat(mService.onIsLanguageAvailable("eng", "FRA", ""), isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
         assertThat(mService.onIsLanguageAvailable("ine", "", ""), isTtsLangCode(TextToSpeech.LANG_NOT_SUPPORTED));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
     }
 
     public void testOnGetDefaultVoiceNameFor() {
-        assertThat(mService.onLoadLanguage("vie", "VNM", "saigon"), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
+        assertThat(mService.onLoadLanguage("vie", "VNM", "central"), isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
 
-        assertThat(mService.onGetDefaultVoiceNameFor("eng", "", ""), is("en"));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        assertThat(mService.onGetDefaultVoiceNameFor("eng", "", ""), is("en-gb"));
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
         assertThat(mService.onGetDefaultVoiceNameFor("eng", "USA", ""), is("en-us"));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
-        assertThat(mService.onGetDefaultVoiceNameFor("eng", "GBR", "scotland"), is("en-sc"));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        assertThat(mService.onGetDefaultVoiceNameFor("eng", "GBR", "scotland"), is("en-gb-scotland"));
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
         assertThat(mService.onGetDefaultVoiceNameFor("eng", "USA", "rp"), is("en-us"));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
-        assertThat(mService.onGetDefaultVoiceNameFor("eng", "", "scotland"), is("en"));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        assertThat(mService.onGetDefaultVoiceNameFor("eng", "", "scotland"), is("en-gb"));
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
-        assertThat(mService.onGetDefaultVoiceNameFor("eng", "FRA", "rp"), is("en"));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        assertThat(mService.onGetDefaultVoiceNameFor("eng", "FRA", "rp"), is("en-gb"));
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
-        assertThat(mService.onGetDefaultVoiceNameFor("eng", "FRA", ""), is("en"));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        assertThat(mService.onGetDefaultVoiceNameFor("eng", "FRA", ""), is("en-gb"));
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
 
         assertThat(mService.onGetDefaultVoiceNameFor("ine", "", ""), is(nullValue()));
-        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "saigon");
+        checkLanguage(mService.onGetLanguage(), "vie", "VNM", "central");
         assertThat(mService.getActiveVoice(), is(notNullValue()));
-        assertThat(mService.getActiveVoice().name, is("vi-sgn"));
+        assertThat(mService.getActiveVoice().name, is("vi-vn-x-central"));
     }
 
     public void testLanguages() {

android/eSpeakTests/src/com/reecedunn/espeak/test/TextToSpeechTestCase.java

@@ -72,7 +72,7 @@ public class TextToSpeechTestCase extends AndroidTestCase
                 Log.d("TextToSpeechTestCase", "setUp: available = " + available);
             }
 
-            assertThat(available, isTtsLangCode(TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE));
+            assertThat(available, isTtsLangCode(TextToSpeech.LANG_AVAILABLE));
         }
         catch (Exception e)
         {

android/eSpeakTests/src/com/reecedunn/espeak/test/VoiceData.java

@@ -75,90 +75,101 @@ public class VoiceData
     }
 
     public static final Voice[] voices = new Voice[] {
-        new Voice("af",          "other/af",     "af",  "afr", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Afrikaans",                        "afr",              "Hierdie is 'n voorbeeld van gesproke teks in Afrikaans"),
-        new Voice("am",          "test/am",      "am",  "amh", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Amharic",                          "amh",              "ይሄ በአማርኛ ላይ የተነገረ ጽሑፍ ናሙና ነው።", "ይሄ በAmharic ላይ የተነገረ ጽሑፍ ናሙና ነው።"),
-        new Voice("an",          "europe/an",    "an",  "arg", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Aragonese",                        "arg",              "This is a sample of text spoken in Aragonese"),
-        new Voice("as",          "test/as",      "as",  "asm", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Assamese",                         "asm",              "This is a sample of text spoken in Assamese", "This is a sample of text spoken in অসমীয়া"),
-        new Voice("az",          "test/az",      "az",  "aze", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Azerbaijani",                      "aze",              "This is a sample of text spoken in Azerbaijani", "This is a sample of text spoken in azərbaycan"),
-        new Voice("bg",          "europe/bg",    "bg",  "bul", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Bulgarian",                        "bul",              "Това е откъс от изговорен текст на български"),
-        new Voice("bn",          "asia/bn",      "bn",  "ben", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Bengali",                          "ben",              "This is a sample of text spoken in বাংলা", "This is a sample of text spoken in Bengali"),
-        new Voice("bs",          "europe/bs",    "bs",  "bos", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Bosnian",                          "bos",              "This is a sample of text spoken in Bosnian", "This is a sample of text spoken in bosanski"),
-        new Voice("ca",          "europe/ca",    "ca",  "cat", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Catalan",                          "cat",              "Aquesta és una mostra de text llegit en veu alta en català"),
-        new Voice("cs",          "europe/cs",    "cs",  "ces", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Czech",                            "ces",              "Toto je ukázkový text namluvený v jazyce čeština"),
-        new Voice("cy",          "europe/cy",    "cy",  "cym", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Welsh",                            "cym",              "This is a sample of text spoken in Welsh", "This is a sample of text spoken in Cymraeg"),
-        new Voice("da",          "europe/da",    "da",  "dan", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Danish",                           "dan",              "Dette er et eksempel på talt tekst på dansk"),
-        new Voice("de",          "de",           "de",  "deu", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "German",                           "deu",              "Dies ist ein Beispieltext auf Deutsch"),
-        new Voice("el",          "europe/el",    "el",  "ell", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Greek (Modern)",                   "ell",              "Αυτό είναι ένα δείγμα κειμένου που εκφέρεται στα Ελληνικά"),
-        new Voice("en",          "default",      "en",  "eng", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "English",                          "eng",              "This is a sample of text spoken in English"),
-        new Voice("en-sc",       "other/en-sc",  "en",  "eng", "GB",  "GBR", "scotland", SpeechSynthesis.GENDER_MALE,        "English (Scotland)",               "eng-GBR-scotland", "This is a sample of text spoken in English (United Kingdom,Scottish Standard English)"),
-        new Voice("en-gb",       "en",           "en",  "eng", "GB",  "GBR", "",         SpeechSynthesis.GENDER_MALE,        "English (UK)",                     "eng-GBR",          "This is a sample of text spoken in English (United Kingdom)"),
-        new Voice("en-uk-north", "other/en-n",   "en",  "eng", "GB",  "GBR", "north",    SpeechSynthesis.GENDER_MALE,        "English (Lancashire)",             "eng-GBR-north",    "This is a sample of text spoken in English (United Kingdom,NORTH)"),
-        new Voice("en-uk-rp",    "other/en-rp",  "en",  "eng", "GB",  "GBR", "rp",       SpeechSynthesis.GENDER_MALE,        "English (Received Pronunciation)", "eng-GBR-rp",       "This is a sample of text spoken in English (United Kingdom,RP)", "This is a sample of text spoken in English (United Kingdom,rp)"),
-        new Voice("en-uk-wmids", "other/en-wm",  "en",  "eng", "GB",  "GBR", "wmids",    SpeechSynthesis.GENDER_MALE,        "English (West Midlands)",          "eng-GBR-wmids",    "This is a sample of text spoken in English (United Kingdom,WMIDS)"),
-        new Voice("en-us",       "en-us",        "en",  "eng", "US",  "USA", "",         SpeechSynthesis.GENDER_MALE,        "English (US)",                     "eng-USA",          "This is a sample of text spoken in English (United States)"),
-        new Voice("en-wi",       "other/en-wi",  "en",  "eng", "JM",  "JAM", "",         SpeechSynthesis.GENDER_MALE,        "English (Caribbean)",              "eng-JAM",          "This is a sample of text spoken in English (Jamaica)"),
-        new Voice("eo",          "other/eo",     "eo",  "epo", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Esperanto",                        "epo",              "This is a sample of text spoken in Esperanto", "This is a sample of text spoken in esperanto"),
-        new Voice("es",          "europe/es",    "es",  "spa", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Spanish",                          "spa",              "Esto es un ejemplo de texto hablado en español."),
-        new Voice("es-la",       "es-la",        "es",  "spa", "MX",  "MEX", "",         SpeechSynthesis.GENDER_MALE,        "Spanish (Latin America)",          "spa-MEX",          "Esto es un ejemplo de texto hablado en español (México)."),
-        new Voice("et",          "europe/et",    "et",  "est", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Estonian",                         "est",              "This is a sample of text spoken in eesti", "This is a sample of text spoken in Estonian"),
-        new Voice("eu",          "europe/eu",    "eu",  "eus", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Basque",                           "eus",              "This is a sample of text spoken in Basque", "This is a sample of text spoken in euskara"),
-        new Voice("fa",          "asia/fa",      "fa",  "fas", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Farsi (Persian)",                  "fas",              "این یک نمونه از نوشتار گفته شده در فارسی است"),
-        new Voice("fi",          "europe/fi",    "fi",  "fin", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Finnish",                          "fin",              "Tämä on näyte puhutusta tekstistä kielellä suomi"),
-        new Voice("fr-be",       "europe/fr-be", "fr",  "fra", "BE",  "BEL", "",         SpeechSynthesis.GENDER_MALE,        "French (Belgium)",                 "fra-BEL",          "Voici un exemple de texte énoncé en français (Belgique)."),
-        new Voice("fr-fr",       "fr",           "fr",  "fra", "FR",  "FRA", "",         SpeechSynthesis.GENDER_MALE,        "French (France)",                  "fra-FRA",          "Voici un exemple de texte énoncé en français (France)."),
-        new Voice("ga",          "europe/ga",    "ga",  "gle", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Irish",                            "gle",              "This is a sample of text spoken in Irish", "This is a sample of text spoken in Gaeilge"),
-        new Voice("gd",          "test/gd",      "gd",  "gla", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Scottish Gaelic",                  "gla",              "This is a sample of text spoken in Scottish Gaelic", "This is a sample of text spoken in Gàidhlig"),
-        new Voice("grc",         "other/grc",    "grc", "grc", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Greek (Ancient)",                  "grc",              "This is a sample of text spoken in Ancient Greek"),
-        new Voice("gu",          "asia/gu",      "gu",  "guj", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Gujarati",                         "guj",              "This is a sample of text spoken in Gujarati", "This is a sample of text spoken in ગુજરાતી"),
-        new Voice("hi",          "asia/hi",      "hi",  "hin", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Hindi",                            "hin",              "यह हिन्दी में बोले गए पाठ का नमूना है"),
-        new Voice("hr",          "europe/hr",    "hr",  "hrv", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Croatian",                         "hrv",              "Ovo je primjer teksta izgovorenog na hrvatski"),
-        new Voice("hu",          "europe/hu",    "hu",  "hun", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Hungarian",                        "hun",              "Ez egy magyar nyelven felolvasott szöveg mintája."),
-        new Voice("hy",          "asia/hy",      "hy",  "hye", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Armenian",                         "hye",              "This is a sample of text spoken in Armenian", "This is a sample of text spoken in հայերեն"),
-        new Voice("hy-west",     "asia/hy-west", "hy",  "hye", "AM",  "ARM", "arevmda",  SpeechSynthesis.GENDER_MALE,        "Armenian (Western)",               "hye-ARM-arevmda",  "This is a sample of text spoken in Armenian (Armenia,Western Armenian)", "This is a sample of text spoken in հայերեն (Հայաստան,AREVMDA)", "This is a sample of text spoken in հայերեն (Հայաստան,արեւմտահայերէն)"),
-        new Voice("ia",          "other/ia",     "ia",  "ina", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Interlingua",                      "ina",              "This is a sample of text spoken in Interlingua"),
-        new Voice("id",          "asia/id",      "in",  "ind", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Indonesia",                        "ind",              "Ini adalah sebuah contoh teks yang diucapkan di Bahasa Indonesia"), // NOTE: 'id' is the correct ISO 639-1 code, but Android/Java uses 'in'.
-        new Voice("is",          "europe/is",    "is",  "isl", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Icelandic",                        "isl",              "This is a sample of text spoken in íslenska", "This is a sample of text spoken in Icelandic"),
-        new Voice("it",          "europe/it",    "it",  "ita", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Italian",                          "ita",              "Questo è un esempio di testo parlato in italiano"),
-        new Voice("jbo",         "other/jbo",    "jbo", "jbo", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Lojban",                           "jbo",              "This is a sample of text spoken in Lojban"),
-        new Voice("ka",          "asia/ka",      "ka",  "kat", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Georgian",                         "kat",              "This is a sample of text spoken in Georgian", "This is a sample of text spoken in ქართული"),
-        new Voice("kl",          "test/kl",      "kl",  "kal", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Greenlandic",                      "kal",              "This is a sample of text spoken in Kalaallisut", "This is a sample of text spoken in kalaallisut"),
-        new Voice("kn",          "asia/kn",      "kn",  "kan", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Kannada",                          "kan",              "This is a sample of text spoken in Kannada", "This is a sample of text spoken in ಕನ್ನಡ"),
-        new Voice("ko",          "asia/ko",      "ko",  "kor", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Korean",                           "kor",              "한국어로 읽은 텍스트 샘플입니다."),
-        new Voice("ku",          "asia/ku",      "ku",  "kur", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Kurdish",                          "kur",              "This is a sample of text spoken in Kurdish"),
-        new Voice("la",          "other/la",     "la",  "lat", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Latin",                            "lat",              "This is a sample of text spoken in Latin"),
-        new Voice("lt",          "europe/lt",    "lt",  "lit", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Lithuanian",                       "lit",              "Tai teksto, sakomo lietuvių, pavyzdys"),
-        new Voice("lv",          "europe/lv",    "lv",  "lav", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Latvian",                          "lav",              "Šis ir izrunāta teksta paraugs šādā valodā: latviešu."),
-        new Voice("mk",          "europe/mk",    "mk",  "mkd", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Macedonian",                       "mkd",              "This is a sample of text spoken in Macedonian", "This is a sample of text spoken in македонски"),
-        new Voice("ml",          "asia/ml",      "ml",  "mal", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Malayalam",                        "mal",              "This is a sample of text spoken in Malayalam", "This is a sample of text spoken in മലയാളം"),
-        new Voice("mr",          "test/mr",      "mr",  "mar", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Marathi",                          "mar",              "This is a sample of text spoken in Marathi", "This is a sample of text spoken in मराठी"),
-        new Voice("ms",          "asia/ms",      "ms",  "msa", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Malay",                            "msa",              "Ini adalah sampel teks yang dilafazkan dalam Bahasa Melayu", "Ini adalah sampel teks yang dilafazkan dalam Malay"),
-        new Voice("ne",          "asia/ne",      "ne",  "nep", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Nepali",                           "nep",              "This is a sample of text spoken in Nepali", "This is a sample of text spoken in नेपाली"),
-        new Voice("nl",          "europe/nl",    "nl",  "nld", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Dutch",                            "nld",              "Dit is een voorbeeld van tekst die is uitgesproken in het Nederlands"),
-        new Voice("no",          "europe/no",    "nb",  "nob", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Norwegian (Bokmål)",               "nob",              "Dette er et teksteksempel lest opp på norsk bokmål"),
-        new Voice("om",          "test/om",      "om",  "orm", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Oromoo",                           "orm",              "This is a sample of text spoken in Oromo", "This is a sample of text spoken in Oromoo"),
-        new Voice("or",          "test/or",      "or",  "ori", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Oriya",                            "ori",              "This is a sample of text spoken in Oriya", "This is a sample of text spoken in ଓଡ଼ିଆ"),
-        new Voice("pa",          "asia/pa",      "pa",  "pan", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Panjabi",                          "pan",              "This is a sample of text spoken in Punjabi", "This is a sample of text spoken in ਪੰਜਾਬੀ"),
-        new Voice("pap",         "test/pap",     "pap", "pap", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Papiamento",                       "pap",              "This is a sample of text spoken in Papiamento"),
-        new Voice("pl",          "europe/pl",    "pl",  "pol", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Polish",                           "pol",              "To jest przykład tekstu mówionego przy ustawieniu polski"),
-        new Voice("pt-br",       "pt",           "pt",  "por", "BR",  "BRA", "",         SpeechSynthesis.GENDER_MALE,        "Portuguese (Brazil)",              "por-BRA",          "Esta é uma amostra de texto falado em português (Brasil)"),
-        new Voice("pt-pt",       "europe/pt-pt", "pt",  "por", "PT",  "PRT", "",         SpeechSynthesis.GENDER_MALE,        "Portuguese (Portugal)",            "por-PRT",          "Este é um exemplo de texto falado em português (Portugal)"),
-        new Voice("ro",          "europe/ro",    "ro",  "ron", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Romanian",                         "ron",              "Aceasta este o mostră de text vorbit în română"),
-        new Voice("ru",          "europe/ru",    "ru",  "rus", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Russian",                          "rus",              "Так синтезатор речи озвучивает русский текст"),
-        new Voice("si",          "test/si",      "si",  "sin", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Sinhalese",                        "sin",              "This is a sample of text spoken in Sinhala", "This is a sample of text spoken in සිංහල"),
-        new Voice("sk",          "europe/sk",    "sk",  "slk", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Slovak",                           "slk",              "Toto je ukážkový text nahovorený v jazyku slovenčina"),
-        new Voice("sl",          "test/sl",      "sl",  "slv", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Slovenian",                        "slv",              "To je vzorec besedila, izgovorjen v slovenščina"),
-        new Voice("sq",          "europe/sq",    "sq",  "sqi", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Albanian",                         "sqi",              "This is a sample of text spoken in Albanian", "This is a sample of text spoken in Shqip", "This is a sample of text spoken in shqip"),
-        new Voice("sr",          "europe/sr",    "sr",  "srp", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Serbian",                          "srp",              "Ово је пример текста који је изговорен на језику Српски", "Ово је пример текста који је изговорен на језику српски"),
-        new Voice("sv",          "europe/sv",    "sv",  "swe", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Swedish",                          "swe",              "Detta är ett textexempel som läses på svenska"),
-        new Voice("sw",          "other/sw",     "sw",  "swa", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Swahili",                          "swa",              "Hii ni sampuli ya maandishi yaliyonenwa katika Kiswahili", "Hii ni sampuli ya maandishi yaliyonenwa katika Swahili"),
-        new Voice("ta",          "asia/ta",      "ta",  "tam", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Tamil",                            "tam",              "This is a sample of text spoken in தமிழ்", "This is a sample of text spoken in Tamil"),
-        new Voice("te",          "asia/te",      "te",  "tel", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Telugu",                           "tel",              "This is a sample of text spoken in Telugu", "This is a sample of text spoken in తెలుగు"),
-        new Voice("tr",          "asia/tr",      "tr",  "tur", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Turkish",                          "tur",              "Bu, Türkçe dilinde seslendirilen örnek bir metindir"),
-        new Voice("ur",          "test/ur",      "ur",  "urd", "",    "",    "",         SpeechSynthesis.GENDER_UNSPECIFIED, "Urdu",                             "urd",              "This is a sample of text spoken in Urdu", "This is a sample of text spoken in اردو"),
-        new Voice("vi",          "asia/vi",      "vi",  "vie", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Vietnamese",                       "vie",              "Đây là mẫu văn bản được đọc bằng Tiếng Việt"),
-        new Voice("vi-hue",      "asia/vi-hue",  "vi",  "vie", "VN",  "VNM", "hue",      SpeechSynthesis.GENDER_MALE,        "Vietnamese",                       "vie-VNM-hue",      "Đây là mẫu văn bản được đọc bằng Tiếng Việt (Việt Nam,HUE)", "Đây là mẫu văn bản được đọc bằng Tiếng Việt (Việt Nam,hue)"),
-        new Voice("vi-sgn",      "asia/vi-sgn",  "vi",  "vie", "VN",  "VNM", "saigon",   SpeechSynthesis.GENDER_MALE,        "Vietnamese",                       "vie-VNM-saigon",   "Đây là mẫu văn bản được đọc bằng Tiếng Việt (Việt Nam,SAIGON)"),
-        new Voice("zh",          "asia/zh",      "zh",  "zho", "",    "",    "",         SpeechSynthesis.GENDER_MALE,        "Chinese (Mandarin)",               "zho",              "This is a sample of text spoken in 中文"),
-        new Voice("zh-yue",      "asia/zh-yue",  "zh",  "zho", "HK",  "HKG", "",         SpeechSynthesis.GENDER_MALE,        "Chinese (Cantonese)",              "zho-HKG",          "This is a sample of text spoken in 中文 (中華人民共和國香港特別行政區)", "This is a sample of text spoken in 中文 (香港)"),
+        new Voice("af",              "gmw/af",              "af",  "afr", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Afrikaans",                        "afr",              "Hierdie is 'n voorbeeld van gesproke teks in Afrikaans"),
+        new Voice("am",              "sem/am",              "am",  "amh", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Amharic",                          "amh",              "ይሄ በአማርኛ ላይ የተነገረ ጽሑፍ ናሙና ነው።", "ይሄ በAmharic ላይ የተነገረ ጽሑፍ ናሙና ነው።"),
+        new Voice("an",              "roa/an",              "an",  "arg", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Aragonese",                        "arg",              "This is a sample of text spoken in Aragonese"),
+        new Voice("ar",              "sem/ar",              "ar",  "ara", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Arabic",                           "ara",              "هذه عينة من النص الذي تم نطقه في العربية"),
+        new Voice("as",              "inc/as",              "as",  "asm", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Assamese",                         "asm",              "This is a sample of text spoken in Assamese", "This is a sample of text spoken in অসমীয়া"),
+        new Voice("az",              "trk/az",              "az",  "aze", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Azerbaijani",                      "aze",              "This is a sample of text spoken in Azerbaijani", "This is a sample of text spoken in azərbaycan"),
+        new Voice("bg",              "zls/bg",              "bg",  "bul", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Bulgarian",                        "bul",              "Това е откъс от изговорен текст на български"),
+        new Voice("bn",              "inc/bn",              "bn",  "ben", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Bengali",                          "ben",              "This is a sample of text spoken in বাংলা", "This is a sample of text spoken in Bengali"),
+        new Voice("bs",              "zls/bs",              "bs",  "bos", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Bosnian",                          "bos",              "This is a sample of text spoken in Bosnian", "This is a sample of text spoken in bosanski"),
+        new Voice("ca",              "roa/ca",              "ca",  "cat", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Catalan",                          "cat",              "Aquesta és una mostra de text llegit en veu alta en català"),
+        new Voice("cs",              "zlw/cs",              "cs",  "ces", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Czech",                            "ces",              "Toto je ukázkový text namluvený v jazyce čeština"),
+        new Voice("cy",              "cel/cy",              "cy",  "cym", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Welsh",                            "cym",              "This is a sample of text spoken in Welsh", "This is a sample of text spoken in Cymraeg"),
+        new Voice("da",              "gmq/da",              "da",  "dan", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Danish",                           "dan",              "Dette er et eksempel på talt tekst på dansk"),
+        new Voice("de",              "gmw/de",              "de",  "deu", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "German",                           "deu",              "Dies ist ein Beispieltext auf Deutsch"),
+        new Voice("el",              "grk/el",              "el",  "ell", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Greek (Modern)",                   "ell",              "Αυτό είναι ένα δείγμα κειμένου που εκφέρεται στα Ελληνικά"),
+        new Voice("en-gb-scotland",  "gmw/en-GB-scotland",  "en",  "eng", "GB",  "GBR", "scotland", SpeechSynthesis.GENDER_MALE, "English (Scotland)",               "eng-GBR-scotland", "This is a sample of text spoken in English (United Kingdom,Scottish Standard English)"),
+        new Voice("en-gb",           "gmw/en",              "en",  "eng", "GB",  "GBR", "",         SpeechSynthesis.GENDER_MALE, "English (UK)",                     "eng-GBR",          "This is a sample of text spoken in English (United Kingdom)"),
+        new Voice("en-gb-x-gbclan",  "gmw/en-GB-x-gbclan",  "en",  "eng", "GB",  "GBR", "gbclan",   SpeechSynthesis.GENDER_MALE, "English (Lancashire)",             "eng-GBR-gbclan",   "This is a sample of text spoken in English (United Kingdom,GBCLAN)"),
+        new Voice("en-gb-x-rp",      "gmw/en-GB-x-rp",      "en",  "eng", "GB",  "GBR", "rp",       SpeechSynthesis.GENDER_MALE, "English (Received Pronunciation)", "eng-GBR-rp",       "This is a sample of text spoken in English (United Kingdom,RP)", "This is a sample of text spoken in English (United Kingdom,rp)"),
+        new Voice("en-gb-x-gbcwmd",  "gmw/en-GB-x-gbcwmd",  "en",  "eng", "GB",  "GBR", "gbcwmd",   SpeechSynthesis.GENDER_MALE, "English (West Midlands)",          "eng-GBR-gbcwmd",   "This is a sample of text spoken in English (United Kingdom,GBCWMD)"),
+        new Voice("en-us",           "gmw/en-US",           "en",  "eng", "US",  "USA", "",         SpeechSynthesis.GENDER_MALE, "English (US)",                     "eng-USA",          "This is a sample of text spoken in English (United States)"),
+        new Voice("en-029",          "gmw/en-029",          "en",  "eng", "JM",  "JAM", "",         SpeechSynthesis.GENDER_MALE, "English (Caribbean)",              "eng-JAM",          "This is a sample of text spoken in English (Jamaica)"),
+        new Voice("eo",              "art/eo",              "eo",  "epo", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Esperanto",                        "epo",              "This is a sample of text spoken in Esperanto", "This is a sample of text spoken in esperanto"),
+        new Voice("es",              "roa/es",              "es",  "spa", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Spanish",                          "spa",              "Esto es un ejemplo de texto hablado en español."),
+        new Voice("es-419",          "roa/es-419",          "es",  "spa", "MX",  "MEX", "",         SpeechSynthesis.GENDER_MALE, "Spanish (Latin America)",          "spa-MEX",          "Esto es un ejemplo de texto hablado en español (México)."),
+        new Voice("et",              "urj/et",              "et",  "est", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Estonian",                         "est",              "This is a sample of text spoken in eesti", "This is a sample of text spoken in Estonian"),
+        new Voice("eu",              "eu",                  "eu",  "eus", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Basque",                           "eus",              "This is a sample of text spoken in Basque", "This is a sample of text spoken in euskara"),
+        new Voice("fa",              "ira/fa",              "fa",  "fas", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Farsi (Persian)",                  "fas",              "این یک نمونه از نوشتار گفته شده در فارسی است"),
+        new Voice("fi",              "urj/fi",              "fi",  "fin", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Finnish",                          "fin",              "Tämä on näyte puhutusta tekstistä kielellä suomi"),
+        new Voice("fr-be",           "roa/fr-BE",           "fr",  "fra", "BE",  "BEL", "",         SpeechSynthesis.GENDER_MALE, "French (Belgium)",                 "fra-BEL",          "Voici un exemple de texte énoncé en français (Belgique)."),
+        new Voice("fr-ch",           "roa/fr-CH",           "fr",  "fra", "CH",  "CHE", "",         SpeechSynthesis.GENDER_MALE, "French (Switzerland)",             "fra-CHE",          "Voici un exemple de texte énoncé en français (Suisse)."),
+        new Voice("fr-fr",           "roa/fr",              "fr",  "fra", "FR",  "FRA", "",         SpeechSynthesis.GENDER_MALE, "French (France)",                  "fra-FRA",          "Voici un exemple de texte énoncé en français (France)."),
+        new Voice("ga",              "cel/ga",              "ga",  "gle", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Irish",                            "gle",              "This is a sample of text spoken in Irish", "This is a sample of text spoken in Gaeilge"),
+        new Voice("gd",              "cel/gd",              "gd",  "gla", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Scottish Gaelic",                  "gla",              "This is a sample of text spoken in Scottish Gaelic", "This is a sample of text spoken in Gàidhlig"),
+        new Voice("gn",              "sai/gn",              "gn",  "grn", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Guarani",                          "grn",              "This is a sample of text spoken in Guarani"),
+        new Voice("grc",             "grk/grc",             "grc", "grc", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Greek (Ancient)",                  "grc",              "This is a sample of text spoken in Ancient Greek"),
+        new Voice("gu",              "inc/gu",              "gu",  "guj", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Gujarati",                         "guj",              "This is a sample of text spoken in Gujarati", "This is a sample of text spoken in ગુજરાતી"),
+        new Voice("hi",              "inc/hi",              "hi",  "hin", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Hindi",                            "hin",              "यह हिन्दी में बोले गए पाठ का नमूना है"),
+        new Voice("hr",              "zls/hr",              "hr",  "hrv", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Croatian",                         "hrv",              "Ovo je primjer teksta izgovorenog na hrvatski"),
+        new Voice("hu",              "urj/hu",              "hu",  "hun", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Hungarian",                        "hun",              "Ez egy magyar nyelven felolvasott szöveg mintája."),
+        new Voice("hy",              "ine/hy",              "hy",  "hye", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Armenian",                         "hye",              "This is a sample of text spoken in Armenian", "This is a sample of text spoken in հայերեն"),
+        new Voice("hy-arevmda",      "ine/hy-arevmda",      "hy",  "hye", "AM",  "ARM", "arevmda",  SpeechSynthesis.GENDER_MALE, "Armenian (Western)",               "hye-ARM-arevmda",  "This is a sample of text spoken in Armenian (Armenia,Western Armenian)", "This is a sample of text spoken in հայերեն (Հայաստան,AREVMDA)", "This is a sample of text spoken in հայերեն (Հայաստան,արեւմտահայերէն)"),
+        new Voice("ia",              "art/ia",              "ia",  "ina", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Interlingua",                      "ina",              "This is a sample of text spoken in Interlingua"),
+        new Voice("id",              "poz/id",              "in",  "ind", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Indonesia",                        "ind",              "Ini adalah sebuah contoh teks yang diucapkan di Bahasa Indonesia"), // NOTE: 'id' is the correct ISO 639-1 code, but Android/Java uses 'in'.
+        new Voice("is",              "gmq/is",              "is",  "isl", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Icelandic",                        "isl",              "This is a sample of text spoken in íslenska", "This is a sample of text spoken in Icelandic"),
+        new Voice("it",              "roa/it",              "it",  "ita", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Italian",                          "ita",              "Questo è un esempio di testo parlato in italiano"),
+        new Voice("ja",              "jpx/ja",              "ja",  "jpn", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Japanese",                         "jpn",              "日本語で話すテキストサンプルです。"),
+        new Voice("jbo",             "art/jbo",             "jbo", "jbo", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Lojban",                           "jbo",              "This is a sample of text spoken in Lojban"),
+        new Voice("ka",              "ccs/ka",              "ka",  "kat", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Georgian",                         "kat",              "This is a sample of text spoken in Georgian", "This is a sample of text spoken in ქართული"),
+        new Voice("kl",              "esx/kl",              "kl",  "kal", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Greenlandic",                      "kal",              "This is a sample of text spoken in Kalaallisut", "This is a sample of text spoken in kalaallisut"),
+        new Voice("kn",              "dra/kn",              "kn",  "kan", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Kannada",                          "kan",              "This is a sample of text spoken in Kannada", "This is a sample of text spoken in ಕನ್ನಡ"),
+        new Voice("ko",              "ko",                  "ko",  "kor", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Korean",                           "kor",              "한국어로 읽은 텍스트 샘플입니다."),
+        new Voice("kok",             "inc/kok",             "kok", "kok", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Konkani",                          "kok",              "This is a sample of text spoken in कोंकणी"),
+        new Voice("ku",              "ira/ku",              "ku",  "kur", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Kurdish",                          "kur",              "This is a sample of text spoken in Kurdish"),
+        new Voice("ky",              "trk/ky",              "ky",  "kir", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Kyrgyz",                           "kir",              "This is a sample of text spoken in кыргызча"),
+        new Voice("la",              "itc/la",              "la",  "lat", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Latin",                            "lat",              "This is a sample of text spoken in Latin"),
+        new Voice("lt",              "bat/lt",              "lt",  "lit", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Lithuanian",                       "lit",              "Tai teksto, sakomo lietuvių, pavyzdys"),
+        new Voice("lv",              "bat/lv",              "lv",  "lav", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Latvian",                          "lav",              "Šis ir izrunāta teksta paraugs šādā valodā: latviešu."),
+        new Voice("mi",              "poz/mi",              "mi",  "mri", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Māori",                            "mri",              "This is a sample of text spoken in Maori"),
+        new Voice("mk",              "zls/mk",              "mk",  "mkd", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Macedonian",                       "mkd",              "This is a sample of text spoken in Macedonian", "This is a sample of text spoken in македонски"),
+        new Voice("ml",              "dra/ml",              "ml",  "mal", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Malayalam",                        "mal",              "This is a sample of text spoken in Malayalam", "This is a sample of text spoken in മലയാളം"),
+        new Voice("mr",              "inc/mr",              "mr",  "mar", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Marathi",                          "mar",              "This is a sample of text spoken in Marathi", "This is a sample of text spoken in मराठी"),
+        new Voice("ms",              "poz/ms",              "ms",  "msa", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Malay",                            "msa",              "Ini adalah sampel teks yang dilafazkan dalam Bahasa Melayu", "Ini adalah sampel teks yang dilafazkan dalam Malay"),
+        new Voice("mt",              "sem/mt",              "mt",  "mlt", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Maltese",                          "mlt",              "This is a sample of text spoken in Malti"),
+        new Voice("my",              "sit/my",              "my",  "mya", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Myanmar",                          "mya",              "This is a sample of text spoken in Telugu", "This is a sample of text spoken in ဗမာ"),
+        new Voice("ne",              "inc/ne",              "ne",  "nep", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Nepali",                           "nep",              "This is a sample of text spoken in Nepali", "This is a sample of text spoken in नेपाली"),
+        new Voice("nl",              "gmw/nl",              "nl",  "nld", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Dutch",                            "nld",              "Dit is een voorbeeld van tekst die is uitgesproken in het Nederlands"),
+        new Voice("nb",              "gmq/nb",              "nb",  "nob", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Norwegian (Bokmål)",               "nob",              "Dette er et teksteksempel lest opp på norsk bokmål"),
+        new Voice("om",              "cus/om",              "om",  "orm", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Oromoo",                           "orm",              "This is a sample of text spoken in Oromo", "This is a sample of text spoken in Oromoo"),
+        new Voice("or",              "inc/or",              "or",  "ori", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Oriya",                            "ori",              "This is a sample of text spoken in Oriya", "This is a sample of text spoken in ଓଡ଼ିଆ"),
+        new Voice("pa",              "inc/pa",              "pa",  "pan", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Panjabi",                          "pan",              "This is a sample of text spoken in Punjabi", "This is a sample of text spoken in ਪੰਜਾਬੀ"),
+        new Voice("pap",             "roa/pap",             "pap", "pap", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Papiamento",                       "pap",              "This is a sample of text spoken in Papiamento"),
+        new Voice("pl",              "zlw/pl",              "pl",  "pol", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Polish",                           "pol",              "To jest przykład tekstu mówionego przy ustawieniu polski"),
+        new Voice("pt",              "roa/pt",              "pt",  "por", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Portuguese",                       "por",              "Este é um exemplo de texto falado em português"),
+        new Voice("pt-br",           "roa/pt-BR",           "pt",  "por", "BR",  "BRA", "",         SpeechSynthesis.GENDER_MALE, "Portuguese (Brazil)",              "por-BRA",          "Esta é uma amostra de texto falado em português (Brasil)"),
+        new Voice("ro",              "roa/ro",              "ro",  "ron", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Romanian",                         "ron",              "Aceasta este o mostră de text vorbit în română"),
+        new Voice("ru",              "zle/ru",              "ru",  "rus", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Russian",                          "rus",              "Так синтезатор речи озвучивает русский текст"),
+        new Voice("sd",              "inc/sd",              "sd",  "snd", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Sindhi",                           "snd",              "This is a sample of text spoken in Sindhi"),
+        new Voice("si",              "inc/si",              "si",  "sin", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Sinhalese",                        "sin",              "This is a sample of text spoken in Sinhala", "This is a sample of text spoken in සිංහල"),
+        new Voice("sk",              "zlw/sk",              "sk",  "slk", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Slovak",                           "slk",              "Toto je ukážkový text nahovorený v jazyku slovenčina"),
+        new Voice("sl",              "zls/sl",              "sl",  "slv", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Slovenian",                        "slv",              "To je vzorec besedila, izgovorjen v slovenščina"),
+        new Voice("sq",              "ine/sq",              "sq",  "sqi", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Albanian",                         "sqi",              "This is a sample of text spoken in Albanian", "This is a sample of text spoken in Shqip", "This is a sample of text spoken in shqip"),
+        new Voice("sr",              "zls/sr",              "sr",  "srp", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Serbian",                          "srp",              "Ово је пример текста који је изговорен на језику Српски", "Ово је пример текста који је изговорен на језику српски"),
+        new Voice("sv",              "gmq/sv",              "sv",  "swe", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Swedish",                          "swe",              "Detta är ett textexempel som läses på svenska"),
+        new Voice("sw",              "bnt/sw",              "sw",  "swa", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Swahili",                          "swa",              "Hii ni sampuli ya maandishi yaliyonenwa katika Kiswahili", "Hii ni sampuli ya maandishi yaliyonenwa katika Swahili"),
+        new Voice("ta",              "dra/ta",              "ta",  "tam", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Tamil",                            "tam",              "This is a sample of text spoken in தமிழ்", "This is a sample of text spoken in Tamil"),
+        new Voice("te",              "dra/te",              "te",  "tel", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Telugu",                           "tel",              "This is a sample of text spoken in Telugu", "This is a sample of text spoken in తెలుగు"),
+        new Voice("tn",              "bnt/tn",              "tn",  "tsn", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Setswana",                         "tsn",              "This is a sample of text spoken in Tswana"),
+        new Voice("tr",              "trk/tr",              "tr",  "tur", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Turkish",                          "tur",              "Bu, Türkçe dilinde seslendirilen örnek bir metindir"),
+        new Voice("tt",              "trk/tt",              "tt",  "tat", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Tatar",                            "tat",              "This is a sample of text spoken in Tatar"),
+        new Voice("ur",              "inc/ur",              "ur",  "urd", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Urdu",                             "urd",              "This is a sample of text spoken in Urdu", "This is a sample of text spoken in اردو"),
+        new Voice("vi",              "aav/vi",              "vi",  "vie", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Vietnamese",                       "vie",              "Đây là mẫu văn bản được đọc bằng Tiếng Việt"),
+        new Voice("vi-vn-x-central", "aav/vi-VN-x-central", "vi",  "vie", "VN",  "VNM", "central",  SpeechSynthesis.GENDER_MALE, "Vietnamese",                       "vie-VNM-central",  "Đây là mẫu văn bản được đọc bằng Tiếng Việt (Việt Nam,CENTRAL)", "Đây là mẫu văn bản được đọc bằng Tiếng Việt (Việt Nam,central)"),
+        new Voice("vi-vn-x-south",   "aav/vi-VN-x-south",   "vi",  "vie", "VN",  "VNM", "south",    SpeechSynthesis.GENDER_MALE, "Vietnamese",                       "vie-VNM-south",    "Đây là mẫu văn bản được đọc bằng Tiếng Việt (Việt Nam,SOUTH)"),
+        new Voice("cmn",             "sit/cmn",             "zh",  "zho", "",    "",    "",         SpeechSynthesis.GENDER_MALE, "Chinese (Mandarin)",               "zho",              "This is a sample of text spoken in 中文"),
+        new Voice("yue",             "sit/yue",             "zh",  "zho", "HK",  "HKG", "",         SpeechSynthesis.GENDER_MALE, "Chinese (Cantonese)",              "zho-HKG",          "This is a sample of text spoken in 中文 (中華人民共和國香港特別行政區)", "This is a sample of text spoken in 中文 (香港)"),
     };
 }

android/jni/Android.mk

@@ -5,7 +5,7 @@ LOCAL_CFLAGS = -std=c11
 
 # ucd-tools wide-character compatibility support:
 
-UCDTOOLS_SRC_PATH  := ../../ucd-tools/src
+UCDTOOLS_SRC_PATH  := ../../src/ucd-tools/src
 UCDTOOLS_SRC_FILES := \
   $(subst $(LOCAL_PATH)/$(UCDTOOLS_SRC_PATH),$(UCDTOOLS_SRC_PATH),$(wildcard $(LOCAL_PATH)/$(UCDTOOLS_SRC_PATH)/*.c*))
 
@@ -33,7 +33,6 @@ ESPEAK_SOURCES := \
   src/libespeak-ng/synthdata.c \
   src/libespeak-ng/synthesize.c \
   src/libespeak-ng/synth_mbrola.c \
-  src/libespeak-ng/tokenizer.c \
   src/libespeak-ng/translate.c \
   src/libespeak-ng/tr_languages.c \
   src/libespeak-ng/voices.c \

android/jni/Application.mk

@@ -1,4 +1,4 @@
 APP_OPTIM             := release
-APP_PLATFORM          := android-8
+APP_PLATFORM          := android-14
 APP_TOOLCHAIN_VERSION := 4.6.3
 APP_ABI               := mips armeabi armeabi-v7a x86

android/jni/include/config.h

@@ -1,2 +1,2 @@
 // This is created by autoconf on POSIX-compatibe build environments.
-#define PACKAGE_VERSION "1.49.2-dev"
+#define PACKAGE_VERSION "1.49.3-dev"

android/src/com/reecedunn/espeak/SpeechSynthesis.java

@@ -114,6 +114,8 @@ public class SpeechSynthesis {
                 return new Locale(parts[0], parts[1]);
             case 3: // language-country-variant
                 return new Locale(parts[0], parts[1], parts[2]);
+            case 4: // language-country-x-privateuse
+                return new Locale(parts[0], parts[1], parts[3]);
             default:
                 return null;
         }
@@ -355,6 +357,7 @@ public class SpeechSynthesis {
     static {
         mJavaToIanaLanguageCode.put("afr", "af");
         mJavaToIanaLanguageCode.put("amh", "am");
+        mJavaToIanaLanguageCode.put("ara", "ar");
         mJavaToIanaLanguageCode.put("arg", "an");
         mJavaToIanaLanguageCode.put("asm", "as");
         mJavaToIanaLanguageCode.put("aze", "az");
@@ -377,6 +380,7 @@ public class SpeechSynthesis {
         mJavaToIanaLanguageCode.put("fra", "fr");
         mJavaToIanaLanguageCode.put("gle", "ga");
         mJavaToIanaLanguageCode.put("gla", "gd");
+        mJavaToIanaLanguageCode.put("grn", "gn");
         mJavaToIanaLanguageCode.put("guj", "gu");
         mJavaToIanaLanguageCode.put("hin", "hi");
         mJavaToIanaLanguageCode.put("hrv", "hr");
@@ -386,9 +390,11 @@ public class SpeechSynthesis {
         mJavaToIanaLanguageCode.put("ind", "in"); // NOTE: The deprecated 'in' code is used by Java/Android.
         mJavaToIanaLanguageCode.put("isl", "is");
         mJavaToIanaLanguageCode.put("ita", "it");
+        mJavaToIanaLanguageCode.put("jpn", "ja");
         mJavaToIanaLanguageCode.put("kat", "ka");
         mJavaToIanaLanguageCode.put("kal", "kl");
         mJavaToIanaLanguageCode.put("kan", "kn");
+        mJavaToIanaLanguageCode.put("kir", "ky");
         mJavaToIanaLanguageCode.put("kor", "ko");
         mJavaToIanaLanguageCode.put("kur", "ku");
         mJavaToIanaLanguageCode.put("lat", "la");
@@ -397,7 +403,10 @@ public class SpeechSynthesis {
         mJavaToIanaLanguageCode.put("mkd", "mk");
         mJavaToIanaLanguageCode.put("mal", "ml");
         mJavaToIanaLanguageCode.put("mar", "mr");
+        mJavaToIanaLanguageCode.put("mlt", "mt");
+        mJavaToIanaLanguageCode.put("mri", "mi");
         mJavaToIanaLanguageCode.put("msa", "ms");
+        mJavaToIanaLanguageCode.put("mya", "my");
         mJavaToIanaLanguageCode.put("nep", "ne");
         mJavaToIanaLanguageCode.put("nld", "nl");
         mJavaToIanaLanguageCode.put("nob", "nb");
@@ -412,12 +421,15 @@ public class SpeechSynthesis {
         mJavaToIanaLanguageCode.put("sin", "si");
         mJavaToIanaLanguageCode.put("slk", "sk");
         mJavaToIanaLanguageCode.put("slv", "sl");
+        mJavaToIanaLanguageCode.put("snd", "sd");
         mJavaToIanaLanguageCode.put("sqi", "sq");
         mJavaToIanaLanguageCode.put("srp", "sr");
         mJavaToIanaLanguageCode.put("swe", "sv");
         mJavaToIanaLanguageCode.put("swa", "sw");
         mJavaToIanaLanguageCode.put("tam", "ta");
         mJavaToIanaLanguageCode.put("tel", "te");
+        mJavaToIanaLanguageCode.put("tat", "tt");
+        mJavaToIanaLanguageCode.put("tsn", "tn");
         mJavaToIanaLanguageCode.put("tur", "tr");
         mJavaToIanaLanguageCode.put("urd", "ur");
         mJavaToIanaLanguageCode.put("vie", "vi");
@@ -426,6 +438,7 @@ public class SpeechSynthesis {
         mJavaToIanaCountryCode.put("ARM", "AM");
         mJavaToIanaCountryCode.put("BEL", "BE");
         mJavaToIanaCountryCode.put("BRA", "BR");
+        mJavaToIanaCountryCode.put("CHE", "CH");
         mJavaToIanaCountryCode.put("FRA", "FR");
         mJavaToIanaCountryCode.put("GBR", "GB");
         mJavaToIanaCountryCode.put("HKG", "HK");
@@ -435,18 +448,11 @@ public class SpeechSynthesis {
         mJavaToIanaCountryCode.put("USA", "US");
         mJavaToIanaCountryCode.put("VNM", "VN");
 
-        // Map eSpeak locales to their correct BCP47 locales supported by Android:
-        mLocaleFixes.put("en-sc", new Locale("en", "GB", "scotland"));
-        mLocaleFixes.put("en-uk-north", new Locale("en", "GB", "north"));
-        mLocaleFixes.put("en-uk-rp", new Locale("en", "GB", "rp"));
-        mLocaleFixes.put("en-uk-wmids", new Locale("en", "GB", "wmids"));
-        mLocaleFixes.put("en-wi", new Locale("en", "JM"));
-        mLocaleFixes.put("es-la", new Locale("es", "MX"));
-        mLocaleFixes.put("fa-pin", null); // Script tags not supported.
-        mLocaleFixes.put("hy-west", new Locale("hy", "AM", "arevmda")); // hy-arevmda crashes on Android 5.0
-        mLocaleFixes.put("no", new Locale("nb")); // 'no' is valid, but the system uses the more specific 'nb' and 'nn'
-        mLocaleFixes.put("vi-hue", new Locale("vi", "VN", "hue"));
-        mLocaleFixes.put("vi-sgn", new Locale("vi", "VN", "saigon"));
-        mLocaleFixes.put("zh-yue", new Locale("zh", "HK"));
+        // Fix up BCP47 locales not handled correctly by Android:
+        mLocaleFixes.put("cmn", new Locale("zh"));
+        mLocaleFixes.put("en-029", new Locale("en", "JM"));
+        mLocaleFixes.put("es-419", new Locale("es", "MX"));
+        mLocaleFixes.put("hy-arevmda", new Locale("hy", "AM", "arevmda")); // hy-arevmda crashes on Android 5.0
+        mLocaleFixes.put("yue", new Locale("zh", "HK"));
     }
 }

configure.ac

@@ -1,5 +1,5 @@
 AC_PREREQ([2.63])
-AC_INIT([eSpeak NG], [1.49.2-dev], [https://github.com/espeak-ng/espeak-ng/issues], [espeak-ng], [https://github.com/espeak-ng/espeak-ng])
+AC_INIT([eSpeak NG], [1.49.3-dev], [https://github.com/espeak-ng/espeak-ng/issues], [espeak-ng], [https://github.com/espeak-ng/espeak-ng])
 AM_INIT_AUTOMAKE()
 LT_INIT
 
@@ -83,6 +83,23 @@ else
 	AC_MSG_ERROR([C99 is not supported by $CC.])
 fi
 
+dnl ================================================================
+dnl FreeBSD check.
+dnl ================================================================
+AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
+		   #if ! defined(__FreeBSD__)
+		   #error macro not defined
+		   #endif
+		   ]])], [FREEBSD_FOUND="yes"], [FREEBSD_FOUND="no"])
+
+AC_MSG_CHECKING([if targeting FreeBSD])
+if test "$FREEBSD_FOUND" = "yes" ; then
+	AC_MSG_RESULT([-D__BSD_VISIBLE])
+	CFLAGS="$CFLAGS -D__BSD_VISIBLE"
+else
+	AC_MSG_RESULT([no])
+fi
+
 dnl ================================================================
 dnl library checks.
 dnl ================================================================
@@ -252,6 +269,19 @@ AM_CONDITIONAL(HAVE_RU_EXTENDED_DICTIONARY,  [test x"$have_extdict_ru"  = xyes])
 AM_CONDITIONAL(HAVE_ZH_EXTENDED_DICTIONARY,  [test x"$have_extdict_zh"  = xyes])
 AM_CONDITIONAL(HAVE_ZHY_EXTENDED_DICTIONARY, [test x"$have_extdict_zhy" = xyes])
 
+dnl ================================================================
+dnl Compiler warnings.
+dnl
+dnl Reference: https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html
+dnl ================================================================
+
+AX_CHECK_COMPILE_FLAG([-Wimplicit],              [CFLAGS="-Wimplicit $CFLAGS"])
+AX_CHECK_COMPILE_FLAG([-Wmissing-prototypes],    [CFLAGS="-Wmissing-prototypes $CFLAGS"])
+AX_CHECK_COMPILE_FLAG([-Wreturn-type],           [CFLAGS="-Wreturn-type $CFLAGS"])
+AX_CHECK_COMPILE_FLAG([-Wuninitialized],         [CFLAGS="-Wuninitialized $CFLAGS"])
+AX_CHECK_COMPILE_FLAG([-Wunused],                [CFLAGS="-Wunused $CFLAGS"])
+AX_CHECK_COMPILE_FLAG([-Wunused-parameter],      [CFLAGS="-Wunused-parameter $CFLAGS"])
+
 dnl ================================================================
 dnl Generate output.
 dnl ================================================================

dictsource/af_list

@@ -218,7 +218,9 @@ afganistan  afg'anistan
 alaska      al'aska
 albanië     alb'A:ne@:@-
 algerië     alx2'e@re@:@-
+alhambra    al'ambra
 amanzimtoti $4 
+amstel      amst@l
 antwerpen   antv&rp@n
 avignon     _^_FR
 babilon     babilOn
@@ -333,6 +335,7 @@ oberammergau o@b@r'am@rgaU
 oklahoma    @Ukl@h'@Uma
 outeniekwa  @Ut@n'ikwa
 oxford      _^_EN
+paardeneiland pA:rd@n_'eIlant 
 palermo     pal'&rmu
 pelindaba   p&l@nd'A:ba
 perú        p@ru
@@ -342,6 +345,7 @@ portugal    pOrtyx2al
 potchefstroom pOtSIfstr'o@m
 rhône       _^_FR
 richardsbaai ritS@dsb'AI
+riversdal   r@v@rsdal
 riviersonderend r@fi:rsOn@r_'Ent
 robertson   _^_EN
 salvador    salvadO:r
@@ -380,7 +384,7 @@ uruguay     urugwaI
 wakloof     vA:klo@f
 zimbabwe    zI:mb'ab_wE
 zulu        zulu // Kwazulu-Natal
-
+zürich      _^_DE
 
 // People names
 abdullah    abd'ula
@@ -585,6 +589,7 @@ james       _^_EN
 jane        _^_EN
 	janine      dZ@ni:n	
 Jean        ZA~n $capital
+jeff        _^_EN
 jesebel     je@s@b&l
 jessica     _^_EN
 jimmy       _^_EN
@@ -603,6 +608,7 @@ julia       _^_EN
 jung        juN
 khayyam     kaj'am
 kelvin      _^_EN
+kennedy     _^_EN
 kim         _^_EN $capital
 kleopatra   kliu:p'A:tra
 krige       krix2@
@@ -769,9 +775,11 @@ stockenström stOk@nstro@m
 stuart      _^_EN
 suzanne     suz'A:n
 suzette     suz'Et
+sylvia      _^_EN
 tania       tanja
 telemann    te@l@man
 terblanche  t@rblA:nS
+thelma      _^_EN
 theo        tiu
 theron tr'On
 thessalonicense tEsalo@nis'E:ns@
@@ -816,6 +824,7 @@ vivaldi     viv'aldi
 viviers     v@v@je@
 wilhelm     v@lh&l@-m
 williston   _^_EN
+woltemade   vOlt@mA:d@
 khumalo     kum'A:lu
 zuma        zu:ma
 
@@ -825,6 +834,7 @@ alibama     $3
 antares     $2
 beatles     _^_EN
 boeing      _^_EN
+cadillac    _^_EN
 checkers    tSEk@rs
 chevrolet   _^_FR
 chrysler    kraIsl@r
@@ -1044,7 +1054,8 @@ fides         _^_LA
 forma         _^_LA
 grata         _^_LA
 habitatio     _^_LA
-inclusio      _^_LA
+inclusio      _^_LA
+(in camera)   @n||kam@ra
 (in debiti)   _^_LA
 (in absentia) _^_LA	
 (in extremis) _^_LA	
@@ -1075,7 +1086,9 @@ vivos         _^_LA
 // main word list
 
 aangaande     $2
+aanmerklik    $2
 aanstaande    $2
+aanswel       A:nsw&l
 (a cappella)  a||kap'&la
 adagio        ad'A:dZi;%@U
 afgaanse      afg'A:ns@
@@ -1091,6 +1104,7 @@ algaande      $2
 alge          alx2@
 algehele      alx2@h,e@l@
 allegro       al'Egru
+allengs       alENs
 allergene     $3
 almiskie      $3
 alom          al_'Om
@@ -1116,11 +1130,13 @@ aversie       av'&rsi
 
 babelaas      bab@lA:s
 barrikade     $3
+beaming       b@_A:m@N
 bedewete      be@d@ve@t@	
 beide         beId@
 bekaf         b&kaf
 belangriker   b@laNr@k,Ir
 bene          be@n@
+beringde      b@rINd@
 beswil        bEsv@l	
 besnedene     b@sne@d@n@	
 bestes        bEst@s
@@ -1130,6 +1146,7 @@ bewe          be@v@
 bewebeen      be@v@be@n
 bewend        be@v@nt
 bewering      b@ve@rIN
+bilharzia     b@lharsia
 biopsie       bi'Opsi
 bomaat      bo@mA:t
 bordegoed     bO:rd@x2ut
@@ -1155,6 +1172,7 @@ cedille       s@dIl@
 charisma      kar'Isma
 cinsaut       s@nso@
 cliché 	kliS'eI:
+clientèle     _^_FR
 clivia        klIvija
 cognac        kOn^ak
 confetti      $2
@@ -1173,6 +1191,7 @@ dawidster     dA:v@tst&r
 déjà          _^_FR
 dekade        dEk'A:d@
 dekreling     dEkre@l@N
+demensie      d@me~nsi
 deurentyd     dy@r@nteIt
 deurgaans     $1
 deurkruis     $2
@@ -1183,6 +1202,7 @@ diffuus       d@fy:s
 dikwels       dIkv@ls
 diplomasie    d@plo@mas'i
 disleksie     $2
+dissiplinering d@s@plin'e@r@N
 divan         div'an
 doktore       $2
 doyen       dO:j'En
@@ -1219,12 +1239,14 @@ ewentwil      e@v@ntv@l
 exodus        Eks'o@dWs
 
 factotum      $2
+faktotum      $2
 fakture       $2
 fetakaas      fEtakA:s
 figuur        f@x2yr
 filippense    f@l@pEns@
 finalis       $3
 finaliste     $3
+fluktuasie    $3
 fort          fOrt
 forte         fOrt@
 
@@ -1269,6 +1291,7 @@ hekke         h&k@
 helaas        he@l'A:s
 here          he@r@
 herero        hEr'E:ru
+herontmoeting h&r_Ontmut@N
 herrysenis    h&r'eIs@n@s
 hierso        hi:rsO
 hippie        _^_EN
@@ -1294,8 +1317,10 @@ inkatha       iNk'A:ta
 inkluis       $2
 innestel      InnEst@l
 insomnia      @nsOmnija
+inteling      Inte@l@N
 intens @nt'Ens
 intensiteit   $4
+inwegery      Inve@x2@reI
 ironieë       irun'i:@
 
 jantwak       jantw'ak
@@ -1306,12 +1331,14 @@ kafee         kaf'e@
 kaliefa       kal'ifa
 kameraderie   kam@rA:d@r'i
 kapittel      kap'It@l
+karolie       kar_o@li
 kateter       kat'e@t@r
 katswink      katsv@nk
 kaviaar       kavi'A:r
 kennisvaardig $1
 kimono        $2
 klaasvakie    $2
+kliënteel     $3
 klimeid       klImeIt
 knapsekêrel   knaps@k&:r@l
 kolossense    kOl@s'e~ns@
@@ -1325,11 +1352,13 @@ kopiereg      kup'ir&x2
 korswel       kOrsv@l
 kotiljons     kOt@lj'o~ns
 kritiek       krIt'ik
+kruisteling   krYyste@l@N
 kulture       $2
 kunsmatige    kWnsm'A:t@x2@
 kwansuis      $2
 kweekwal      kwe@kval
 
+landswye      lantsveI@
 lasagne       las'anj@
 legaat        l@x2A:t
 legate        l@x2A:t@
@@ -1337,6 +1366,7 @@ libido        l@bidu
 liefdespel    lifd@sp&l
 lipstiek      lIpstik
 lisensiaat    l@se~nsi;'A:t
+logieserwys   lo@x2is@rveIs
 lokaas        lOk_A:s
 londense      lOnd@ns@
 lord          _^_EN
@@ -1448,6 +1478,7 @@ oorsee        $2
 oorsese       $2
 oorval        $2 $verb
 opend         Op_Ent
+opwindends    Opv'Ind@nts
 oraal         o@r'A:l
 orale         o@r'A:l@
 oranjepers    o@r'anj@p&:rs
@@ -1472,6 +1503,7 @@ pinotage      pinut'A:Z
 pirouette     p@ru:'Et
 plaasjaap     plA:sjA:p
 plasenta      plas'Enta
+platwaai      platvAI
 poinsettia    pOIns'Etia
 pond          po~nt
 poniepers     po@nip&rs
@@ -1504,6 +1536,7 @@ salute        sal'yt@
 samelewing    sA:m@l,e@v@N
 sanhedrin     sanh'e@dr@n
 sangemeenskap sanx2@me@nskap
+sedeles       se@d@lEs
 sedertdien    $3
 sektore       $2
 sinagoge      s@n@x2o@x2@
@@ -1532,6 +1565,7 @@ sonore        sun'o@r@
 sonure        sOn_yr@
 sover         so@f&r
 staccato      $2
+steekhoudendheid $2
 sterwens      st&rv@ns
 stilswye      $1
 strydros      streIt_rOs
@@ -1554,7 +1588,10 @@ tesourus      t@s@UrWs
 tevrede       t@fre@d@
 thula         tu:la
 toegee        tux2e@
+toegeklik     tux2@klIk
+toemessel     tumEs@l
 toereken      ture@k@n
+toleransie    tOl@r'ansi
 tornado       $2
 totale        tut'A:l@
 totsiens      $2
@@ -1566,6 +1603,7 @@ tsoenami      tsun'A:mi
 uitdaging     YydA:x2@N
 uiteraard     Yyt@r_'A:rt
 uitermate     $3
+uitgeslotene  Yytx2@slo@t@n@
 uniforme      $3
 
 vaarwel       fA:rv'&l
@@ -1622,6 +1660,7 @@ voëlent       fo@@l_Ent
 vraprys       frA:preIs
 vrek          fr&k
 vroeërig      fru:@r@x2
+vrygelatene   freIx2@lA:t@n@
 vu            _^_FR
 
 waarskynlik   $2
@@ -1638,6 +1677,7 @@ welaf         v&l_af
 weleer        v&l_'e@r
 wentaktiek    $1
 willens       vIl@ns
+wydoog        veIt_o@x2
 wysgere       veIsx2e@r@
 
 êrenster      &:r@nst@r

dictsource/af_rules

@@ -73,6 +73,7 @@
  @@@smokkel) ary          %ar%eI // default stress: diamant-/drank-/dwelm-/goud-/kokaïensmokkelary, etc.
              amarula      %am%arul%a // fix stress and a sounds: amarula and compounds
              a (ment      %a // shorten a sound: perkament/testament and compounds
+          _) arendag (CAC %A:r@ntax2 // fix e sound: arendagtig/-e/-heid   
              arends       A:r@nts_ // fix e sound: arendsoog/-kloue/-nes
              argen (tA    %arx2@n // move default stress: Argentinië/Argentyns/-e
              a (riA       'A: // akwarium/barium/estuarium/herbarium/seminaria
@@ -310,7 +311,7 @@
              attaché      _%at%aSe@ // correct pronunciation: (handels-/inligtings-)attaché
              attrib (u    %atr@b // move default stress: attribuut/bute/attributêr
           K) au           @U // trauma/-ties/Aucamp/Paul/-a/-us
-             auer         aU@r //Sauer etc. fixed au and e sounds in compounds.
+             auer         aU@r //Sauer etc. fix au and e sounds in compounds.
              augustus     %Ox2WstWs
              avokado      %af%ukA:du // stress and o sounds
              avokade      %af%ukA:d@ // variant form of avokado
@@ -431,7 +432,8 @@
              bo (grond    bo@ // fix 1st o sound: bogronds/-e
           _) bo (kle      bo@ // fix o sound: bokleed/boklere
              bokma (kier  b%Okm%a // fix stress and a sound: bokmakierie/-s/-tjie
-             bom (AA      bOm_ // fix a sound, pause: bomaanval/-aanslag/-aard/-eenheid and compounds
+             bom (AA      bOm_ // fix a sound, pause: bomaanval/-aard/-eenheid and compounds
+             bomaans (la  bOm_A:ns // fix o sound: bomaanslag/-aanslae
              bomaans      bo@mA:ns // but fix O sound: bomaans/-e
              bonde (C     bOnd@ // verbondenheid/bondeldraer/gebondene/saambondelend
              boos (aardig b%o@s_ // move default stress: boosaardig/-e/-heid
@@ -526,6 +528,7 @@
              bakate (l    b%ak@t& // fix stress and vowel sounds: bakatel/-le/-letjie
              baken        bA:k@n // fix e sound: afbakening and compounds
              bakte (ri    b%akte@ // fix stress and e sound: bakterie/-ë
+             balalaika    b%al%alaIk%a // fix a sounds and stress: balalaika/-s/-musiek
           _) bam (boes    b%am // move default stress: bamboes/-e/bamboesfluit...
              ba (nalA     b%a // fix stress and 1st a sound: banale/banaliteit
           band (Alier     b,and // fix stress and d sound: band(e/o)lier/-e
@@ -547,6 +550,7 @@
              ba (sieli    b%a // fix stress and 1st a sound: basielie/-kruit, etc.
              basotho      b%asut%u // fix stress and vowel sounds: Basotho/-0ponie and similar
              basi (s      bA:s@ // basis and compounds
+             basilie      b%asil%i // fix a sound and stress: basilie/-kruid, etc.
              bastille     b%asti:l // Bastille and compounds like Bastilledag
              batal (jon   b%at%al // shorten 1st a sound, moved stress: bataljon and compounds
              ba (tik      b%a // fix stress and a sound: batik/-doek/-kuns/-werk
@@ -646,6 +650,8 @@
              bere (_      be@r@ //tediebere pandabere etc.
           _) ber (C       b&r // fix e sound: Bert/Berta/Bertie/Bertus/berke/-boom
              beste (_     bEst@ // fix e sound: beste/allerbeste/naasbeste/-s
+             beton (C@   b@tOn // split ng sounds: betongebou/-gietsel/-gruis
+             beton (inC@  b@t'On // fix O sound: betoninrigting/-ingenieur
              be (weging   b@ // draaibeweging/swaaibeweging
              bewende      be@v@nd@ // first e pronounced long
              bewe (rig    be@v@ // fix e sound and stress: bewerig/-e/-heid
@@ -749,6 +755,7 @@
              dia (C       d%i%a // diabeet/diafragma/diagnose/dialek/dialoog/diamant
              diaken       d%iA:k@n // diaken and compounds
              diako (nie   d%i%ak%u // fix stress and o sound: diakonie/-ë
+             diende       dind@ // fix last e sound in many compounds of bediende: bediendekamer
              diens (willi d%ins // move default stress: dienswillig/-e/-heid
              dieper (@    dip@r // fix e sound: diepere/dieperliggend/-e
              digi (ta     d%ix2%i // move default stress: digitaal/digitale
@@ -983,6 +990,7 @@
          @C) e (reekK     @ // fix e sound: duine-/rotsereeks
         aai) e (C         @ // fix e sound: baaierd/waaierstert/paaiement/compounds starting with baaiers-
         lat) eres         @r@s //lateres
+         @C) erm (K       &r@-m // a catch-all for words ending in -erm(s)
          fp) ers (_       &rs // fix e sound: dof-/gif-/olyf-/tydskrifpers, etc.
       bloup) ers          &:rs // fix e sound: bloupers
         iew) ers (ter     @rs // fix e sound: (l)iewerster
@@ -1074,6 +1082,7 @@
           p) e (talje     @ // fix stress and 1st e sound: petalje and compounds
           l) e (moen      @ //fix lemoen and compounds
          Cy) e (C+        @ // fix connecting e sound in compounds like byekorf/skilderyemuseum, etc.
+             effe (kleur  Ef@ // fix stress and 2nd e sound: effekleur(ig/-e)
              eier         eI@r // eiergeel/eierwit/leierfiguur/leiergroep
          @C) ei (land     _,eI // insert short pause: skiereiland and many -eiland compounds
           @) ei (sen      _'eI // (on)veeleisend/spoedeisend
@@ -1451,6 +1460,7 @@
              een (Cjie    e@:iN // eendjie/Leentjie/steentjies
              een (vorm    %e@n // move default stress: eenvormig/-e/-heid
           _) een (voud@   %e@n // stress on 2nd slb: eenvoudig, but on 1st: eenvoud
+             eend (agtig  %e@nt_ // fix d sound, insert pause: eendagtig/-e/-heid
              eerbied (A   e@rb'id // eerbiedig/ing
              eerbiedw     %e@rb%itv // eerbiedwaardig/eerbiedwekkend
           _) eerw         %e@rv // stress: eerwaarde
@@ -1589,6 +1599,7 @@
              flu (we      fl%y // move default stress: fluweel/fluwele and compounds
           _) fok (o       fOk_ // fix o sound, insert short break: fokop/fokof
              folio        fo@li_u // fix o sounds: folio and compounds
+             fondsw       fOntsv // fix v sound: fondswerwing & compounds/-waardes
              fondue       f%Ondy // fix stress and ue sound: fondue and compounds
              fone (tiek   f%o@n@ // fix e sound: fonetiek and compounds
              fone (tie    f%une@ // fix stress and o sound: foneties/-e
@@ -1602,6 +1613,7 @@
              formi (da    f%Orm%i // move default stress: formidabel/-e
              formu (lier  f%Orm%y // move default stress: formulier/-e/-boek
           _) for (se_     f'Or // stress back to 1st slb; se rule: _) for (C 
+             forse (nd    fOrs@ // fix e sound: forsend/-e
              fos (fa      f%Os // move default stress: fosfaat/fosfate
              fo (ssiel    f%O // move default stress: fossiel/-e and compounds
              foto         fo@tu
@@ -1694,6 +1706,7 @@
           _) gra (na      x2r@ // granaat(boom)/granate
              grandi (o    x2r%and%i // move default stress: grandioos/-ose
           _) gra (vA      x2r%a // fix stress and a sound: gravin/-ne/gravure
+             grenadella   gr@n@d&la // fix stress and vowels: grenadella and compounds like -heining
              griekwa      x2rikwa // fix w sound and a sound in compounds: Griekwa/-land, etc.
        ouCa) gr (ootjie   x2r' // move default stress: ouma/oupagrootjie/-s
           _) graad (e     x2r%A:t_ // fix stress and d sound: graadeen(tjies)/-elfs
@@ -1726,6 +1739,7 @@
              gegesel      x2@x2e@s@l // fix middle e sound
           K) gele (_      x2e@l@ // fix stress and e sound: (eier)gele
           _) geler        x2e@l@r // stress and 1st e sound: geler/gelerig
+          _) gell         x2&l
              gemel (d     x2@m&l // fix e sound: bo-/bowe-/laas-/meergemeld/-e
              gems         x2Ems // fix e sound: (baster)gemsbok/-bul/-ooi, etc.
              gene (_      x2e@n@ // gene/diegene
@@ -1741,7 +1755,7 @@
           _) ge (ossP2    x2@ // but ge- prefix: geossilleer/geossifiseer, etc.
          ni) ge (ri       x2'e@ //nigeriese etc.
              gese (_      x2'e@s@ //Portugese, and others
-
+             gesp (etjie  x2Esp // fix e sounds: gespetjie/-s
              gespe (_     x2Esp@ // gespe and compounds
              gespes (_    x2Esp@s // gespes and compounds
              gewens (g    x2e@v@ns // (on)vergewensgesind/-e/-heid
@@ -1847,6 +1861,7 @@
              hart (stogte_  h%art // move default stress: hartstogtelik
              hart (stogte_N hart // restore default stress: hartstogte
              ha (we       hA: // fix stress and a sound: hawearbeider/lewendehaweafdeling
+          _) hef (a       hEf_ // fix e sound, insert break: hefapparaat/-arm(s)
              hia (sint    h%ij%a // fix stress: hiasint/-e and compounds
           _) hi (bis      h%i // move default stress: hibiskus/-se and compounds
              hierna (maal h%i:rnA: // fix stress and a sound: hiernamaals/-e
@@ -1984,6 +1999,7 @@
           _) idi (o       %id%i // idioom/idiome/idioot
           _) id (A        %id // idille/idillies/ideëryk
          &l) iker (C      @k@r // fix i sound: menslikerwys/redelikerwyse, etc.
+         &l) iker (_      @k@r // fix i sound: afstootliker/(ge)redeliker/onberispeliker
              illumi (nA   %il%um%i // fix stress and vowel sounds: illuminasie/illumineer/illuminati
              illu (si     %ily // illusie/-s/illusief
              illu (strA   %il%W // fix i sound: illustreer/illustrering/illustrasie
@@ -2114,6 +2130,7 @@
           _) in (a@P2     In // inakkuraat/inaktief/inaktiwiteit
           _) inbe (lC     Inb%& // fix e sound in compounds like inbelprogram
              indone (si   @nd%uni: // fix stress and vowel sounds: Indonesië/Indonesiese
+          s) in (gestel   @n // split n g: compounds with ingesteldheid/winsingestelde
           _) in (oe       In_ // fix i sound, pause: inoefen/-oes and derivatives
              invest (eer  @nv%Est // fix stress, v and e sounds: investeer/-der
              investe (r@  @nv%Este@ // fix e sound: (kapitaal)investering/investerende
@@ -2269,7 +2286,7 @@
              klaarblyklik klA:rbl'eIkl@k // klaarblyklik/e
              kla (kous    klA: // fix stress and long a sound: klakous/-e
              klande (sti  kl%and@ // move default stress: klandestien/-e
-          _) kla (sA@@    kla // klasaantekeninge/klasonderwyser and other compounds
+          _) klas (A@     klas_ // klasaantekeninge/klasonderwyser/-opdrag(te)/-ure and other compounds
              kla (ssiek   kl%a // move default stress: klassiek/-e	
              klavesimbel  klA:v@s@mb@l // fix v and e sounds: klavesimbel and compounds
              kla (vier    kl%a // klavier and many compounds
@@ -2324,10 +2341,11 @@
              krieketw     krik@tv //krieket followed by w in compounds always v
 
              krokodi (l   kr%Ok@dI // (wyfie)krokodil/-le
+             kro (niek    kr%u // fix stress and o sound: kroniek and many compounds
              kruger       kr'Y@@r
              ku (ba@      k%y // move default stress: kubaan/kubane
              ku (biek     k%y // move default stress: kubiek/-e/-getal
-          _) kuber        kyb@r // fixed e sound: kuberruim(te)
+          _) kuber        kyb@r // fix e sound: kuberruim(te)
              kulin        k%Wl%in // Move default stress and fix u sound: kulinër/-e
           @) kundi (g     k'Wnd@ // wiskundige/onoordeelkundigheid and many similar
              kurwe        kWrv@ // fix e sound: kurwes/skurwebas/skurwebek/skurwepadda
@@ -2359,11 +2377,13 @@
 
              kabine (t    k%ab%inE // kabinet/drankkabinet/-te/kabinetsvergadering
           _) ka (d@       k%a // stress and 1st a: kadawer-/kadet-/kado/kadot(jie)/kadriel,etc.
+             kadet (A     k%adEt_ // fix e sound, insert pause: kadetorkes/-offisier/-oefening, etc.
              ka (der      kA: // restore stress on long a: kader/-s and compounds
              kad (mium    kad // fix stress and d sound: kadmium and compounds
              kafe (te     k%af@ // fix stress and vowel sounds: kafeteria and compounds
              kafe (ï      k%af%i // fix stress and vowel sounds: kafeïene and compounds
              ka (jak      k%a // fix stress and 1st a sound: kajak/-ke/-vaarder
+             kakao        k%akA:w // fix stress and vowel sounds: kakao and many compounds
              kake (C      kA:k@ // kakebeen/skakelaar/skakelbord/skakelfunksie
              kalahari     kalah'A:ri // stress: Kalahari/-sand/-woestyn
           _) ka (lAnC     k%a // kalender and compounds/kalant/kalander and compounds
@@ -2544,6 +2564,7 @@
              kontrasep    k%Ontr%asEp // fix stress and e sound: kontrasepsie and derivatives
              kontrover (s k%Ontr%uv&r // o and v sounds: kontroversie/kontroversieel
              kop (o       kOp? // fix o sound: koponderstebo/kopomdraai/kopoperasie/gryskoponderwyser/poenskopolifant
+             kop (agtig   k%Op_ // fix o sound, insert break: hamer-/spinnekop-/penkopagtig(e(s))
              kopu (lA     k%Op%y // fix o sound: kopulasie/kopuleer and derivatives
           _) kor (dA      k%Or // kordaat/kordon
              ko (rint     k%u // fix stress and o sound: korint/-e and compounds
@@ -2656,6 +2677,7 @@
              loboto (mie  l%ub%Ot%u // fix stress and o sounds: lobotomie
              lo (ja       l%u // fix stress and o sound: lojale/lojaliteit
              lo (kalA     l%u // fix stress and o sound: lokale/ontvangslokale, etc.
+             loke (t      l%ukE // fix stress and vowel sounds: many compounds with loket
              lom (bardA   l%Om // move default stress: Lombardies/-e/Lombarde
           _) lore (C@     lo@r@ // fix e sound: verloregaan/verloregoederekantoor/Verlorerivier
           _) losge (@P5   l'Osx2@
@@ -2738,6 +2760,7 @@
              mando (lien  m%and%u // fix stress and o sound: mandolien and compounds
              mane (l      m%an& // fix stress and e sound: (kerk)manel(le)punte, etc.
              mane (wale   m%an@ // fix stress and shorten a sound: manewale/-s and compounds
+             maneuvers    m%anY@f@rs // fix last e sound: (krygs)maneuvers, etc.
           _) mango        maNg%u // fix g and o sounds: mangoboord/-sap/-'s
              man (haft    m%an // move default stress: manhaftig/-e/-heid
              mani (akK    mA:n%i // fix stress and 1st a sound: maniak/-ke and compounds
@@ -2805,6 +2828,7 @@
              medisyne     m@d@seIn@ //medisyne and compounds
              meganies     m@x2'A:nis
           _) meege (@P5   m'e@x2@
+          _) meegewe (nd  me@x2e@v@ // fix e sounds and stress: meegewend(e)
              me (juf      m@ // move default stress and shorten e sound
              me (laats    m@ // fix stress and e sound: melaats/-e/-heid
              melancholie (_N m%El%aNk%o@li // fix stress and 1st e sound: melancholie
@@ -2946,6 +2970,7 @@
              morf (otomie m%Orf // move default stress: morfotomie
           _) morr (i      mOr // restore default stress: morrig/morrie/-doring
           _) mors (@      mOrs // restore default stress: morsaf/morsdood/morsig
+             mos (agtig   mOs_ // fix o sound, insert break: (kos)mosagtig(e)
              mosam (biek  m%o@s%am // move default stress: Mosambiek/-er/-se
              mosa (ïek    m%o@s%a // move default stress: mosaïek and compounds
              mo (skee     m%O // move default stress: moskee/-s and compounds
@@ -3035,7 +3060,8 @@
              ne (anderCa  n%i // move default stress: Neander(d/t)al/-ler
              neger (in    n%e@x2@r // move default stress: negerin/-ne
              ne (gosie    n@ // fix stress and e sound: negosie/-ware, etc.
-             nek (om      n&k_ // fix e sound, insert pause: nekom(ge)draai
+             nek (_       n&k // catch-all for words ending in -nek: koedoe-/swaannek
+          _) nek (A       n&k_ // fix e sound, insert pause: nekom(ge)draai/-operasie
              nekta (rien  n%Ekt%a // fix stress and a sound: nektarien/-perske, etc.
              neo (li      n%i%u // fix stress and vowel sounds: neolities/-e/neolitikum
              ner (veu     n%&r // move default stress: nerveus/-e
@@ -3080,6 +3106,7 @@
              nood (lotti  n%o@t // move default stress: noodlottig/-e
              nood (saak   n%o@t // move default stress: noodsaaklik/-e/-heid, vs. noodsaak
              nood (saak_N no@t // restore default stress: (ge)noodsaak
+             nooien (tjie noIN // remove the e sound: nooientjie(s) and compounds
              noord (oos   n%o@rt_ // move default stress: noordoos/-te/-telik/-e
              nor (ma@     n%Or // normaal/normaalweg/normale/abnormaal/-ale
              nostal (gie_N n%Ost%al // stress on last slb.: nostalgie
@@ -3196,6 +3223,7 @@
              ofoon        uf'o@n // mikrofoon/sakso(saxo)foon
              ofone        uf'o@n@ // mikrofone/sakso(saxo)fone
              ograaf       ux2r'A:f // biograaf/demograaf/fotograaf
+             ografeer     ux2raf'e@r // fix o sound: (ge)fotografeer
              ografie      ux2raf'i // biografie/demografie/fotografie
              ografies     ux2r'A:fis // biografies/demografies/fotografies
              ograwe       ux2r'A:v@ // biograwe/demograwe/fotograwe
@@ -3417,6 +3445,7 @@
              ooi          oI
              ooy          oI
              oodjie       oIci
+         CC) ool (A       o@l_ // insert break: skooluur/-ure, steenkooluitvoer, but not: Karoolug
              ootjie       oIci
              oontjie      oINki
              oondjie      oINki
@@ -3483,6 +3512,7 @@
              oot (moedig  %o@t // move default stress: ootmoedig/-e/-heid
 
 .group op
+          _) opaal        %o@pA:l // fix o sound, remove break: opaal and compounds
              opaat        up'A:t // homeopaat/psigopaat and similar
              opatie       upat'i // homeopatie/neuropatie and similar
              opaties      up'A:tis // psigopaties/osteopaties and similar
@@ -3578,6 +3608,7 @@
              pol (vy      p%Ol // move default stress: polvy/-e and compounds
              pomelo       p%ume@l%u // fix stress and o sounds: pomelo(sap/-drankie...)
           _) pon (dok     p%On // move default stress: pondok/-ke/-kie
+             pop (agtig   p%Op_ // fix o sound, insert break: popagtig(e) and compounds
              popu         ,pOpy
           _) por (C       %pOr //portret portaal etc.
              por (ie      p%o@r // move default stress: porie/-ë
@@ -4101,6 +4132,7 @@
              sirene       s@re@n@ // move default stress and fix vowel sounds: sirene/-s/-lied/-sang
              si (troen    s@ // move default stress: sitroen and compounds
              sitronella   s@tr%un&l%a // fix stress and vowel sounds: sitronella and compounds
+             s (juffrou   s // fix s sound: kuns-/kos-/geselskapsjuffrou, etc.
              skade        skA:d@ // elik suffix rule breaks a sound of (on)skadelik
              skadu        skA:dy // skaduagtig/skadutjie
              skaker (ing  sk%ake@r // skakering/-s/kleurskakering
@@ -4118,6 +4150,7 @@
              sker (muts   sk%&r // move default stress: (ge)skermutsel/skermutseling/-e
              skerpi (oen  sk%&rp%i // move default stress: skerpioen/-e and compounds
              skilder (y   sk@ld@r // move stress to y: skildery and compounds like skilderymuseum
+             skim (agtig  sk@m_ // fix i sound, insert break: skimagtig(e)
              skisofr (e   sk%is%ufr // fix stress and o sound: skisofreen/skisofrene
              skle (rose   skl@ // fix stress and e sound: sklerose and compounds
           _) skok (AP4    sk''Ok_ // fix o sound and stress: skokaankondiging/-effek/-insluiting/-onthulling...
@@ -4252,6 +4285,7 @@
           _) su (meri     s%u // fix stress and u sound: sumeries/-e
              su (mmier    s%W // move default stress: sumier/-e
              super (A     s''yp@r_ // fix stress, break in compounds like superintelligent
+             superi (A    s%up%e@r%i // fix stress and vowel sounds: superieur/superioriteit
              superintendent s,upr@nt%EndEnt // fix stress and vowel sounds: superintendent and compounds
              surro (ga    s%Wr%u // fix stress and o sound: surrogaat/surrogate and compounds
              su (saC      s%u // fix stress and u sound: susan/-na/susara
@@ -4271,6 +4305,7 @@
              swepe        swe@p@ // fix w sound: rubberswepe
              sw (erm      sw // fix e sound: naderswerm
              switser      swIts@r // fix e sound: Switserland/Switserse
+             sypel (end   seIp@l // fix e sound: sypelend/-e
 
 .group se
              se (_        s@  //the word se and some ending in se
@@ -4411,6 +4446,7 @@
 
              stede (_     ste@d@ //voorstede, hoofstede etc.
              stele (_     ste@l@ //kastele lepelstele etc.
+             stema (_     ste@ma // fix e sound: doods-/gespreks/-herhalingstema and many similar
              stene (_     ste@n@ //skoorstene, bakstene etc.
              sten (sil    stEn // fix e sound in compounds like wolbaalstensil
        chri) stene (_     st@n@ //fix christene and compounds
@@ -4561,7 +4597,7 @@
              toe (riste@  t%u // move default stress: toeristebedryf/-sentrum and similar
              toer (n      t%ur // move default stress: compounds of toernooi
              toere (_     tu:r@ // restore default stress
-             toi (let     t%OI // move default stress: toilet and compounds
+             toilet       t%OIlEt // move default stress: toilet and compounds: toiletartikel/-emmer/-opsigter
              tokke (lo    t%Ok@ // move default stress: tokkelos/-sie/tokkelok and compounds
              tok (tokk    t%Ok // move default stress: toktokkie/-s/-spelery, etc.
              tombola      t%Ombo@l%a // move default stress: tombola and compounds
@@ -4574,6 +4610,7 @@
           @) toris (_     t'o@r@s // fix stress: pectoris/klitoris
           _) tor (nyn     t%Or // move default stress: tornyn/-e and compounds
           _) tos (ka      t%Os // move default stress: Toskaanse/Toskane
+             tser (tjie   ts@r // fix e sound: (skoen)poetsertjie/weerkaatsertjie
              ttel         t@l // many compounds of bottel/skottelgoed
              tuberkulose  t%yb@rk%ylo@s@ // fix stress; e sound in compounds: tuberkulose/-behandeling
              tug (A       tWx2_ // fix u sound: (on)tugondersoek/-oortreding/-ordonnansie
@@ -4875,6 +4912,7 @@
           _) vanklik      faNkl@k // (on)ontvanklik/-e/-er/-heid
              vanself (spr f%ans%&lf // move default stress: vanselfsprekend/-e/-heid
              vari (A      v%ar%i // fix v sound and stress: variasie/-s/varieer
+             va (sal      v%a // fix stress and v and a sounds: vasal/-le
              vaseline     v%as@lin // fix stress, v and vowel sounds: vaseline/-bottel, etc.
              vasste (l    fast& // fix e sound: vasstel(ling/-lende)
           _) vat (A@      fat_ // fix a sound: vatafstand/-orgaan
@@ -4943,9 +4981,11 @@
 
              ve (l        f& // maagvel, stress on 1st slb
           @) vel (A       fe@l // aanbeveling/aanbevelingsbrief
+             vel (djie    f&l // fix -djie sound: veldjie(s) and compounds
              veld (C      f&lt // fix d sound: veldreuk/-radio/-rantsoen, etc.
              veld (eks    f&lt_ // fix d sound, pause: veldekskursie/-ekspedisie/-s
              veld (o      f&lt_ // fix d sound, pause: veldorgideë/-opsigter/-oppervlakte/-opname, etc.
+             veld (t      f&l // eliminate double t sound: veldtog and many compounds, Langeveldt, Springveldt
              vele         f'e@l@ 
              ven (detta   v%En // fix stress and v sound: vendatta and compounds
              ven (dusie   f@n // fix stress and e sound: vendusie and compounds
@@ -5048,7 +5088,7 @@
              vol (kome    f%Ol
              vo (llA_     fO // volle/vollê/Volla - exception to: vo (lC       f%O
              volle (dig   f%Ole@ // fix stress and e sound: (on)volledig/-e/-heid...
-             voll (engte  fOlE
+             volle (ngte  fOlE
              vo (ller     fO // exception to: vo (lC       f%O
              volles (_    fOl@s // fix stress and e sound: volles/passievolles, etc.
              vol (hou_    fOl // exception to: vo (lC       f%O
@@ -5204,6 +5244,7 @@
              waardeer     vard'e@r // waardeer/waardeerder
              waarder (@   vard'e@r  // waarderend/e/waardering
           &) waardig      v'A:rd@x2 // stress on -waardig: merkwaardig/besienswaardigheid, etc.
+             waar (din    v%A:r // fix stress: waardin/-ne
              waar (neemba v%A:r // move default stress: waarneembaar/-bare
           C) waarts       vA:rts // fix w sound: dakwaarts/huiswaarts/kerkwaarts/noordwaarts/weswaarts
              wêrel (djie  v&:r@l // fix -djie in wêreldjie and compounds

dictsource/ar_list

@@ -251,11 +251,12 @@ $	duulaar
 كغم		kiluu||gHraam 
 طن		t[un
 كم		kiiluu||mitar
-سا 		saaAa 
-م		mitar
-س		su?aal
-ج		dZawaab	
-ص		s[afHah
+سا 		saaAa
+// Units below conflict with single letter name
+// م		mitar 
+// س		su?aal
+// ج		dZawaab	
+// ص		s[afHah
 // العملات
 د.ج     diinaar||dZazaa?irij
 $	    duulaar

dictsource/en_list

@@ -1047,6 +1047,7 @@ autocracy	O:t'0kr@si
 automobile	O:t@m@bi@l
 automaton	O:t'0m@t@n
 autopsy		O:t20psi
+?3 aurora	$2
 avalanche	av@laanS
 avast		$alt3
 avatar		ava#tA@
@@ -1668,7 +1669,7 @@ derogative	$2
 descant		dEskant
 descry		dI#skreI
 desert		dI#z3:t
-desert		dEz3t		$only
+desert		dEz3t		$onlys
 desert		dI#z3:t		$verb
 desi		deIsi
 desolate	dEs@l@t		$only
@@ -2228,7 +2229,7 @@ idevice		$alt6
 idly		aIdlI
 idiocy		IdI@si
 ifrog		$alt6
-ignoramus	Igno@r'eIm@s
+ignoramus	IgnO@r'eIm@s
 illiterate	$alt2
 illumine	$alt2
 imagery		ImIdZri
@@ -2716,6 +2717,7 @@ nonsense	n0ns@ns
 nonetheless	nVnD@l'Es
 nosedive	noUzdaIv
 nosir		noUs3:
+?5 nosir	noUsIR
 not		noUt          // for noted, notable, etc 
 nots		n0ts
 (nôtre dame)	noUtr@'dA:m
@@ -3420,7 +3422,7 @@ sincerest	sIns'i@r@st
 sinus		saIn@s
 siphon		$alt2
 sir		s,3: $only
-?5 sir		s,VR $only
+?5 sir		s,IR $only
 siren		saIr@n
 site		saIt          // for sited
 ski		ski:
@@ -3455,12 +3457,13 @@ sommelier	s0m'Eli@
 son		sVn
 sonar		soUnA@
 sonny		sVnI
+sooth		su:T	$only
+sopapilla	soUp@p'i:@
 sope		soUpeI
+sorbet		sO@beI
 souffle		su:fl'eI
 soundbite	saUndbaIt
 souvenir	su:v@n'i@3
-sopapilla	soUp@p'i:@
-sorbet		sO@beI
 soyabean	sOI@bi:n
 specific	sp@sIfIk
 specimen	spEsI2m@n
@@ -3568,7 +3571,6 @@ tallyho		talI'hoU
 tamale		ta#mA:li
 tampon		tamp0n
 tangerine	tandZ@r'i:n
-taoiseach	ti:S@x
 tapestry	tapI#stri
 tarantula	t@rantS@l@
 tardis		$alt1
@@ -3913,6 +3915,7 @@ yer		j3	$u+
 ye		ji:	$u+
 yea		jeI
 yessir		jEss3:
+?5 yessir	jEssIR
 yoghurt		j0g3t
 ?3 yoghurt	joUg3t
 ?3 yogurt	joUg3t
@@ -4042,6 +4045,7 @@ Assam		A:s'A:m
 Assamese	A:sA:m'i:z
 Audubon		O:d@b0n
 Augusta		$2
+Austen		O2stI2n
 Baghdad		$2
 Bahawalpur	$alt3
 Bahia		ba'i:@
@@ -4546,8 +4550,6 @@ Annise		a#ni:s
 ?!3 Anthony	ant@ni
 Anton		ant0n
 Anya		anj@
-Aoife		i:f@
-Aoiffe		i:f@
 Aphrodite	afr@d'aIti
 Archibald	A@tSIbO:ld
 Archie		A@tSi
@@ -4898,7 +4900,6 @@ Sabine		sa#b'i:n
 Salman		sa#lmA:n
 Samantha	sa#manT@
 (Santa claus)	s'ant@||kl'O:z
-Saoirse		si@S@
 Sarah		se@r@
 Sarisa		$alt3
 Seamus		SeIm@s
@@ -4912,8 +4913,8 @@ Sheila		Si:l@
 Simon		saIm@n
 Sinead		SI2neId
 Sinéad		SI2neId
-Siobhan		S@vO:n
-Siobhán		S@vO:n
+Siobhan		SI2vO:n
+Siobhán		SI2vO:n
 Sonia		s0nj@
 Sophia		soUf'i@
 Sophie		soUfi

dictsource/en_rules

@@ -150,12 +150,16 @@
     sw) a (m_      a
     sw) a (nk      a
         ao         eI0
+    mh) ao         eI		// Irish, e.g. 'mhaol' /weIl/
      m) ao         aU
      p) ao         aU
      t) ao         aU
         ao (_      =aU
-        aois (_    i:S
-        aoise (_   i:S@
+	ao (ism    aU
+	ao (ist    aU
+	aoi (C     i:		// Irish, e.g. 'Aoife' /i:f@/
+	aoir (C    e@		// Irish, e.g. 'Saoirse' /se@S@/
+?3	aoir (C    3:		// Irish, e.g. 'Saoirse' /s3:S@/
      g) ao (l      eI@
         aor        eI'o@
      m) ao (ri     aU
@@ -180,7 +184,8 @@
     _n) a (tional  a
      @) a (tious   'eI
    ell) a (trix    @
-        a (triC    'eI
+        a (trix    'eI
+        a (trice   'eI
      n) a (tur     eI
      n) a (tura    a
      &) a (ture_   @
@@ -512,7 +517,7 @@
     _m) ag (ell    a#dZ
     Cp) age (_     eIdZ
     pp) age (_     I2dZ
-  _ant) ag         'ag
+  _ant) ag (on     'ag
    enr) ag (e_     'eIdZ
   outr) ag (e_     eIdZ
    der) ag (e_     eIdZ
@@ -847,6 +852,7 @@
      _) anony (m   a#n0n@
     _c) an (ood    a#n
      &) an (_      @n
+     &) an (ize    @n
      &) an (_$w_alt1+ an
 ?8   &) an (_      a#n
     @u) an (_S1    n
@@ -1202,6 +1208,7 @@ calend) ar         3
      _) as (in     as
         as (k      aas
      _) as (k      aas
+     s) as (k      as
      _) as (n      az
      _) as (oc     eIs
      C) as (p      aas
@@ -1227,6 +1234,7 @@ calend) ar         3
         ass (u     @S
         ass (ump   a#s
         assur      @SU@
+     _) ass (wh    as
      C) as (t      aas
      C) ast (ell   aas
    _bl) as (to     as  //not aa
@@ -1391,7 +1399,7 @@ _it_separ) ate (_  @t
         au (m_     aU
      _) au (nt     aa
         aur (_     O@
-     _) au (ror    a#
+?!3  _) au (rora   a#
      s) au (rus_   'O:
      @) au (ri     'O:
      C) aus (_     aUs
@@ -1463,7 +1471,7 @@ _it_separ) ate (_  @t
    may) be (_      bi:
      _) be (CA     bI#
      _) bete (lg   bi:t@
-     _) be (C%     bE
+     _) be (C%+    bE
      _) be (atiC   b%i:
      _) be (b      bi:
      _) be (cl     bI#
@@ -1477,7 +1485,7 @@ _it_separ) ate (_  @t
      _) be (kn     bI#
      _) belarus    bEl@r'u:s
      _) be (lC     bE
-     _) be (llig   bI#
+     _) be (llig+  bI#
      _) be (re     bE
      _) be (sC     bI#
      _) be (stia   bE
@@ -2150,6 +2158,8 @@ _it_separ) ate (_  @t
         e (Cical   'E
         e (CiuB    i:
      &) e (_ 
+  aoiC) e (_       @		// Irish, e.g. 'Aoife' /i:fe/
+ aoiCC) e (_       @		// Irish, e.g. 'Saoirse' /se@S@/
    acB) e (_       %I
     XC) e (_N      i:
   vert) e (b       I
@@ -2377,7 +2387,9 @@ _it_separ) ate (_  @t
     @x) ed (_S2vi  d#
      y) ed (_S2v   d#
  debut) ed (_S2    d#
-        edly (_S4m I#dl%i
+        edly (_S3m I#dl%i
+    nt) edly (_S4m I#dl%i
+   tur) edly (_S3v d#l%i
      c) ed (e_     'i:d
      p) ed (e_     i:d
      p) edal       Ed@L
@@ -2567,6 +2579,8 @@ _it_separ) ate (_  @t
         en (core   0n
      &) ency (_    @ns%i
         ency (cli  %EnsI
+     _) en (dg@    %En
+     _) en (dp@    %En
      k) en (d_     En
      s) en (d_     En
      t) en (d_     En
@@ -2771,7 +2785,7 @@ _it_separ) ate (_  @t
    exp) eri (en    i@rI2
     XC) er         3:
     th) er (@      3:
-     h) er (@      %3
+     h) er (nan    %3
      X) er (A      E#r
     _h) eretical   I#rEtIk@L
    _qu) er         3:
@@ -2890,6 +2904,7 @@ _it_separ) ate (_  @t
     &z) es (_S2    %I#z
     &C) es (_S1i   z
    xus) es (_S2    %I#z
+ tamus) es (_S2    %I#z		// hippopotamuses
         es (carp   I2s
         es (cape   %Es
         es (capi   %Es
@@ -3099,7 +3114,7 @@ _it_separ) ate (_  @t
      &) ford (_S4  f3d
     &f) ford (_S4  3d
         for (see   f%O@
-        for (ward  f'o@      // straightforward
+        for (ward  f'O@
         ft (en     f
      &) ful (_S3i  f@L
 
@@ -4288,6 +4303,8 @@ multip) ly         laI
      _) metall (ic m@tal
         metabo     m@t'ab0
      _) meta (llu  m%Eta
+        mh (ao     w		// Irish, e.g. 'mhaol' /weIl/
+   aoi) mh (e      v		// Irish, e.g. 'Caoimhe' /ki:v@/
      &) mobile     m@bi:l
      _) mocha      moUk@
         mono (ga   m@n'0
@@ -4299,6 +4316,7 @@ multip) ly         laI
      &) mouth (_   m@T
      &) mouth (_$w_alt1 maUT
      _) multi      mVlti
+     _) multi (pl  m,VltI
      _) multi (@@P5 m,VltI
 
 .group mi
@@ -5622,6 +5640,7 @@ multip) ly         laI
 
 .group re
      _) re (C@     rI#
+     _) re (collec rE
      _) reflex     ri:flE2ks
      _) reflexive  rI#fl'EksIv
    &Ct) red (_     @d
@@ -5847,6 +5866,10 @@ multip) ly         laI
     Co) s (Er      z
     Co) s (En      z
      e) s (d       z
+   aoi) s (e       S		// Irish, e.g. 'Laoise'
+   aoi) seach      S@x		// Irish, e.g. 'Taoiseach'
+   aoi) sigh       Si		// Irish, e.g. 'Taoisigh'
+  aoir) s (e       S		// Irish, e.g. 'Saoirse'
      _) se (clu    sI#
         secur      sI#kjU@
      _) se (duc    sI#
@@ -5861,6 +5884,7 @@ multip) ly         laI
      @) s (head_   s
      @) s (hill_   s
      @) s (hous    s
+     @) s (hors    s
      _) she'       %Si:
      A) sh (A_     =S
      &) ship (_S4  SIp
@@ -5878,6 +5902,7 @@ multip) ly         laI
      &) s (ic_     z
      n) s (ic_     s
         ss (ic     s
+    mu) s (e       z
     mu) s (ic      z
     ea) s (ie      z
     ea) s (il      z
@@ -6082,6 +6107,8 @@ multip) ly         laI
         th (ill    th
     gh) th (A      th
     ee) thing      DI2N
+   soo) th         D
+   soo) th (say    T
         the (_     D
      &) th (L03_   =T
     ou) thed (_    Dd
@@ -6411,6 +6438,7 @@ contin) u          ju:
      j) ui (c      u:
         uir (_     jU@
      n) ui (s      ju:
+?3   n) ui (s      u:
      r) uis (A     u:z
      r) ui (t      u:
     gr) ui (t      u:I2

dictsource/fa_list

@@ -1,5 +1,5 @@
-// *   Farsi Language fa (or Parsi or Persian) fa_list Version 3.133
-// *   This file writen by Shadyar Khodayari and Ehsan Esmaili who has managed  collecting exceptional words. 05-10-2017
+// *   Farsi Language fa (or Parsi or Persian) fa_list Version 3.134
+// *   This file writen by Shadyar Khodayari and Ehsan Esmaili who has managed  collecting exceptional words. 06-24-2017
 //*********
 // *   This program is free software; you can redistribute it and/or modify  *
 // *   it under the terms of the GNU General Public License as published by  *
@@ -554,6 +554,7 @@ _)	paRAntezbaste:
 آموخت	Amuxt
 آموزد	Amuzad
 آموزش	AmuzeS
+آمپر	AmpeR
 آمپرمتر	AmpeRmetR
 آمپلیفایر	AmpelifAjeR
 آمپیریسم	AmpiRism
@@ -5055,7 +5056,6 @@ _)	paRAntezbaste:
 درایه	deRAje
 درایو	deRAjv
 درایور	deRAjveR
-درباره	daR'bAReje:
 دربازکن	daRbAzkon
 دربدر	daRbedaR
 دربندکشیده	daRbandkeSide
@@ -6809,6 +6809,7 @@ _)	paRAntezbaste:
 شدیدا	Sadidan
 شدیداللحن	Sadidollahn
 شدیم	Sodim
+شراادی	SA:_d:jA:_R
 شرافت	SeRAfat
 شراپنل	SeRApnel
 شراکت	SeRAkat
@@ -6816,7 +6817,6 @@ _)	paRAntezbaste:
 شربت	SaRbat
 شرت	'SoRt
 شرتکات	SoRtkAt
-شراادی	SA:_d:jA:_R
 شرشر	SeRSeR
 شرطه	SoRte:
 شرعا	SaR?an

dictsource/fa_rules

@@ -1,5 +1,5 @@
-// *   Farsi Language fa (or Parsi or Persian) fa_rules Version 3.133
-// *   This file writen by Shadyar Khodayari 05-10-2017
+// *   Farsi Language fa (or Parsi or Persian) fa_rules Version 3.134
+// *   This file writen by Shadyar Khodayari 06-24-2017
 //*********
 // *   This program is free software; you can redistribute it and/or modify  *
 // *   it under the terms of the GNU General Public License as published by  *
@@ -4843,7 +4843,7 @@ L09L04) السّادات (_Sm8	ossAdAt
 			L09L09L09L09) م (L03L09L09L09_	ma
 
 	// Prefixes م
-		_) م (L03L09L03_$noprefixP1@	ma
+		_) م (L03L04L03_$noprefixP1@	ma
 		_) می (L03L09+$noprefixP2@	mi
 		_) می (آL09L09$noprefixP2@	mi
 		_) می (وL09L09$noprefixP2@	mi

dictsource/fi_list

@@ -29,7 +29,7 @@ n	&n
 o	o:
 ö	Y:
 p	pe:
-q	ku:
+q	qu:
 r	&r
 s	&s
 t	te:
@@ -370,22 +370,40 @@ yogurt     jogurt:i
 
 // Abbreviations
 
+(à la)     'ala       $dot    // jonka mukaan
 abc        $abbrev
 adhd       $abbrev
-os         $abbrev    $dot
 ad         $abbrev    $dot    // anno Domini, Herran vuonna
+atk        $abbrev
+ay         $abbrev    // ammaatiyhdistys
 ca         $abbrev    $dot    // circa 
+dna        $abbrev    // elinkeinoelämän keskusliitto
+eg         esimerkiksi $dot   // exempli gratia, esimerkiksi
 em         $abbrev    $dot 
+ek         $abbrev    // elinkeinoelämän keskusliitto
 ekr        $abbrev    $dot    // ennen Kristuksen syntymää
 etc        et_setera  $dot    // et cetera, ym., jne.
-ym         ynn&_mu:ta  $dot   // ynnä muuta (etc)
-eg         esimerkiksi $dot   // exempli gratia, esimerkiksi
+eu         $abbrev
+hiv        $abbrev
+ict        $abbrev
+isbn       $abbrev
+it         $abbrev
 jne        $abbrev    $dot    // ja niin edelleen
-(à la)     'ala       $dot    // jonka mukaan
 ky         $abbrev    $dot    // kommandiittiyhtiö
-oy         $abbrev            // osakeyhtiö 
 nro        numero     $dot
+nvda       $abbrev
+os         $abbrev    $dot
+oy         $abbrev            // osakeyhtiö 
+ry         $abbrev    $dot    // rekisteröity yhdistys
+usb        $abbrev
+utf        $abbrev
+ym         ynn&_mu:ta  $dot   // ynnä muuta (etc)
 
-
-
-
+// foreign words
+blindsquare     _^_EN
+bluetooth     _^_EN
+daisy     _^_EN
+facebook    _^_EN
+talkback     _^_EN
+voiceover     _^_EN
+youtube     _^_EN

dictsource/fi_rules

@@ -165,13 +165,16 @@
 
 
 .group q
-          q          k
-          qq         k:
+          q          q
+          qq         q:
 
 
 .group r
           r          r
-
+      CA) rg (_      Rj // follow swedish pronunciation 
+     _CA) rg (C     Rj  // follow swedish pronunciation 
+          ruoan      ru:an      // Pronounce ruoan as ruuan. Include all inflections and compound words.
+                               // eg. ruoanlaitto, ruoankin, arkiruoan
 
 .group s
           s          s
@@ -197,9 +200,9 @@
 .group u
           u          u
           uu         u:
-          ui         wi
-       _) ui         ui   
-      _j) ui         ui   // [ui] sounds better than [wi] when stressed. Especially <jui> at the beginning of the word sounds bad if pronounced [jwi]
+          ui         ui
+//       _) ui         ui   
+//      _j) ui         ui   // [ui] sounds better than [wi] when stressed. Especially <jui> at the beginning of the word sounds bad if pronounced [jwi]
           uii        u|i:
           uo         uo
           uoo        u|o:

dictsource/it_list

@@ -1,6 +1,6 @@
 
 // This file is UTF-8 encoded
-// Updated 2017 april 28 by Christian Leo M, <[email protected]>
+// Updated 2017  october 3 by Christian Leo M, <[email protected]>
 
 //  $alt   change [e] or [o] in the stressed syllable to [E] or [O]
 //  $alt2  change [E] or [O] in the stressed syllable to [e] or [o]
@@ -102,6 +102,10 @@ _<	min'oRe||d'I
 _>	madZ:'oRe||d'I
 _〈	ap'E@-*ta||,angol'aRe
 _〉	kj'uza||,angol'aRe
+_〈	ap'E@-*ta||angol'ata
+_〉	kj'uza||angol'ata
+_【	ap'E@-*ta||lentikol'aRe
+_】	kj'uza||lentikol'aRe
 *	aste@-*'isko $max3
 $	d'Ol:aRI
 %	pe@-*tS'ENto $max3
@@ -111,7 +115,6 @@ $	d'Ol:aRI
 \	kont@-*ob'a:R*a	$max3
 _|	b'aR*a||ve@-*tik'ale
 &	'a/nd
-©       k'O:pi||raIt
 #	kantSellet:o
 @	ki'otS:ola
 ~	tilde
@@ -137,7 +140,6 @@ _‡	k@-*'otSe||lat'ina||d'Op:ia
 _′	p@-*'imi
 _″	d'Op:io||'apitSe
 _‴	t@-*'iplo||'apitSe
-_‼	esklamatsi'One
 §	setsi'One
 ¶	pa*'ag@-*afo
 _‐	legat'u:Ra
@@ -150,6 +152,8 @@ _‑	t@-*at:,ino||duni'one
 ʔ	glotalst'Op
 ɡ	dZ'i||vel'aRe||ok:luz'iva   // U+0261
 ɣ	g'am:a||min'uskola  // U+0263 voiced velar fricative
+ß	'esse||ak'uta
+ɑ	'alfa
 
 _‣	punt'ato
 ◦	punt'ato
@@ -159,8 +163,6 @@ _‣	punt'ato
 ↓	f@-*'etS:a||_dZ'u
 ♭	bem'Ol:e
 ♯	dI'ezis
-♥	'asso||di||kw'ORI
-❓	p'unto||inteR*ogat'ivo
 _‧	sepaRat'oRe/
 
 ∞	infin'Ito
@@ -787,6 +789,7 @@ black	bl'E:k
 blackout	bl'Ek'aUt
 blind	bl'aInd $onlys
 blues	bl'u:z
+bluff	bl'ef:	$only
 boiler	$1
 bonjour	b,O:_Z'u:r
 brioche	b@-*I'oS
@@ -1020,7 +1023,7 @@ logging	l'Oging
 lord	$alt $only
 love	l'O:v $onlys
 lover	$1
-manche	m'a:nS $only
+manche	m'a:nS $onlys
 my	_^_EN
 myspace	m'aI||sp'Eis
 machine	maS'in	$onlys
@@ -2381,6 +2384,7 @@ ivano	$2
 jace	_^_en $only
 jack	dZ'E:k
 jackson	_^_EN
+jefferson	$1
 jacopo	j'a:kopo
 james	_^_EN
 jane	_^_EN
@@ -2477,6 +2481,7 @@ paul	p'O:l	$only
 peggy	p'Eg:I
 pericle $1
 peter	$1 $onlys
+peterson	p'Ete@-*s,on
 pettarin	$3 $only
 peugeot	p,EZ'o
 piquet	pik'E $only
@@ -3206,6 +3211,7 @@ distino	$1
 distintesi	$2
 disturbano	$2
 disturbino	$2
+dissuadere	dissuad'eRe
 divaghino	$2
 divaricano	$2
 diventino	$2
@@ -4330,6 +4336,7 @@ situino	$1
 sloggiano	$1
 smagnetizzino	$3
 smascher	$1
+smascherano	$1
 smercino	$1
 sniffano	$1
 snodano	$1
@@ -4673,6 +4680,8 @@ zuccherano	$1
 (quando capita)	kw'ando||k'apita
 (se capita)	se||k'a:pita
 (me capita)	me||k'apita
+(cosa capita)	k'oza||k'a:pita
+(come capita)	k'ome||k'a:pita
 (capitano tutte)	k'apitano||t'ut:e
 (che capitano)	ke||k'apitano
 (quando capitano)	kw'ando||k'a:pitano
@@ -4775,6 +4784,7 @@ espellerl	$2
 esprimiti	$2
 evolverci	$2
 evolversi	$2
+facendomici	fatS'endom,itSi
 fammel	$1
 fattel	f'at:el
 fattoci	$1
@@ -4825,7 +4835,6 @@ orientaci	$3
 pensaci	$1
 perdonatel	$3
 permettigli	$2
-pervenutaci	$3
 pizzical	$1
 pizzicami	$1
 portala	$1
@@ -4841,7 +4850,6 @@ pulisciti	$2
 redigerl	$2
 raddrizzarl	@-*ad:@-*its2:'a@-*l
 rafforzal	$2
-regalaci	$2
 regalati	$2	$atstart
 regalavi	$3
 reinseritasi	$4

dictsource/it_listx

@@ -81,6 +81,7 @@ afonie	afon'i;e
 agadez $3
 agapit $2
 agave $1
+agendine	$3
 agenzia $3
 agenzie	adZents'ie
 agerol $2
@@ -137,6 +138,7 @@ allodol $2
 alloggi  $alt
 allora	al:'oRa
 alluce	'al:utSe
+alluci	$1
 almen  $alt2
 alope $1
 alopecia  alopetS'i:a

dictsource/it_rules

@@ -1,7 +1,7 @@
 
 // Italian translation rules
 // This file is UTF-8 encoded
-// Last update: 2017 march   31 by Christian Leo Mameli <[email protected]>
+// Last update: 2017 october 3 by Christian Leo Mameli <[email protected]>
 // letter groups
 // A  any vowel
 // C  any consonant
@@ -28,7 +28,7 @@
 	au	a|%U
 	@C) a (_S1q	a/  // lookup it_list without suffix
 	@i) a (_S1q	a  // lookup it_list without suffix
-	C) a (cY_	'a
+//	C) a (cY_	'a
 	@) a (co_	=a
 	@) a (ca_	=a
 	@) a (che_	=a/
@@ -37,14 +37,15 @@
 	@) a (bile_	'a
 	@) a (ggine_	'a
 	_) ae (r	aE
-	avano (_	'avano
+	avano (_	'avano   // verbs
 	a (tterA_	'a
 	_) all' (P4t	all
 	_) anch' (P5t	ank
 	@) amo (L04_	'amo  // Pron.s verbs
 	Ci) amo (L04_	'amo
 	@) ander (L04_	'andE@-*
-		ate (L04_+	'ate
+	ate (L04_+	'ate
+	abb) assa (L07_	'assa
 	_L04Z) auguri (_	'awguRI
 	astano (_	'astano   // verbs
 
@@ -59,6 +60,7 @@
 	bbano (_	=b:ano // verbs
 	bacia (L07_	b'atSa
 	_) bagna (L07_	b'an^a
+	a) bbraccia (L07_	b:@-*'atS:a
 
 .group c
 	c	k
@@ -90,6 +92,7 @@
 	in) cer (L04_	=tSe@-*
 	cuocer (L04_	kU'OtSe@-*
 	compra (L07_	k'omp@-*a
+	_L04Z) capite (_	kap'ite/
 	_L04Z) capit (A_	k'a:p,it
 	_L04Z) capit (ano_	k'a:p,it
 	correr (L04_	k'o:R*eR
@@ -208,6 +211,8 @@
 	fob (A_	=fob
 	farma (ci_	f'a@-*ma
 	ferma (L07_	f'e@-*ma/
+	_) fatt (AL04_	f'at:
+	_ri) fatt (AL04_	f'at:
 
 .group g
 	g	g
@@ -504,6 +509,7 @@
 	qui (A	kwj
 	_) quell' (P6t	kwEll
 	_) quest' (P6t	kwest
+	_) quant' (P6t	kwant
 
 .group r
 	r	@-*
@@ -522,6 +528,7 @@
 	_) ripeter (L04_	@-*ip'E:te@-*
 	_) romper (L04_	@-*'ompe@-*
 	ca) re (zzaL07_	R'e
+	_) regala (tAL04_	@-*egal'a
 	C) rai (_	@-*'a:I
 	A) rai (_	R'a:I
 	rrai (_	R*'a:I
@@ -563,6 +570,8 @@
 	_) segna (L05_	s'en^a
 	in) segna (L07_ 	s'En^a
 	_)segnala (L07_	sen^'ala
+	specifica (L07_	spetS'ifika
+	stringi (L04_	st@-*'indZi/
 	A) scano (_	=skano  // verbs
 	sciano (_	=Sano
 	A) schiano (_	=skiano
@@ -596,6 +605,7 @@
 	au) tentica (L07_	t'ENtIka/
 	a) ttiva (L07_	t:'iva
 	taglia (L07_	t'al^a
+	tira (L07_	t'iRa
 	L02) tre (_	t@-*'e
 	_) tutt' (P5t	tut:
 
@@ -619,6 +629,7 @@
 	ustano (_	'u:stano
 	C) uo (tano_	U'O
 	uomini (_	w2'Omini
+	fanc) u (lA_	'u:
 
 .group v
 	v	v
@@ -636,6 +647,8 @@
 	_) vota (L07_	v'Ota
 	s) veglia (L07_	v'el^a
 	volta (L07_	v'Olta
+	venut (AL04_	ven'ut
+	_) v' (P2t	v
 
 .group w
 	w	w2
@@ -696,6 +709,7 @@
 	man) zie (_	ts'i;e
 	me) zz	dz:
 	amma) zziamo (L04_	_tsi'a:mo
+	sbara) zz (arL04_	ts:
 
 .group
 

dictsource/ka_emoji

@@ -926,7 +926,8 @@ $textmode
 😖	შეცბუნებული სახე						// [1F616]
 😗	კოცნის გამომხატველი სახე					// [1F617]
 😘	სახე, რომელიც კოცნას აგზავნის					// [1F618]
-😙	კოცნის გამომხატველი სახე მომღიმარე თვალებით			// [1F619]
+😗	kissing face							// [1F617]
+//😙	კოცნის გამომხატველი სახე მომღიმარე თვალებით			// [1F619] TODO: This breaks speaking "7".
 😚	კოცნის გამომხატველი სახე დახუჭული თვალებით			// [1F61A]
 😛	სახე გამოყოფილი ენით						// [1F61B]
 😜	სახე გამოყოფილი ენითა და ჩაკრული თვალით				// [1F61C]
@@ -938,7 +939,8 @@ $textmode
 😢	მტირალა სახე							// [1F622]
 😣	შეუპოვარი სახე							// [1F623]
 😤	სახე ცხვირიდან გამომავალი ორთქლით				// [1F624]
-😥	იმედგაცრუებული, მაგრამ შვების გამომხატველი სახე			// [1F625]
+😥	disappointed but relieved face					// [1F625]
+//😥	იმედგაცრუებული, მაგრამ შვების გამომხატველი სახე			// [1F625] TODO: This breaks speaking "3".
 😦	შეჭმუხნილი სახე ღია პირით					// [1F626]
 😧	ტანჯული სახე							// [1F627]
 😨	შეშინებული სახე							// [1F628]

dictsource/lv_list

@@ -423,6 +423,7 @@ tikpat    $2
 tikvien   $u2+
 turpat    $u2+
 turpretī  $u2
+uzreiz    $2
 vienalga  $2
 vismaz    $2
 

dictsource/lv_rules

@@ -38,6 +38,7 @@
 .L17  ai ais am ā ām ās ie iem o os                // suffixes of certain adjectives finishing with ..ējais
 .L18  am āk i u ot ēt os                           // suffixes of adverbs (apstākļa vārdi)
 
+.L10  ~ at ie iz jā jā jāat jāie jāiz jāno jāsa ne no no pa pār pie pie sa vis // optional verb prefixes
 .L20  ~ ne                                         // optional negation prefix
 
 // +---------------------------+
@@ -62,7 +63,6 @@
 // +---------------------------+
 // |     Pronunciation of o    |
 // +---------------------------+
-.L30  ō ȯ                              // uo pre/post fixes. If facultative writing is used, o is spelled as uo
 
 // Default rules for o with succeeding consonants
 .L40  č ģ ķ ļ ņ ŗ š ž                  // uo indicators — o is spelled as uo in words with Latvian soft consonants
@@ -204,7 +204,7 @@
 
     // wide e ------------------------------------------------------------------------ //
                                      // widening consonants follow
-           e (L25              E
+           e (L25<             E
            e (L25L25           E
                                      // widening vowels follow
            e (L25L24+          E
@@ -253,6 +253,7 @@
      _L20) e (sat_+          _!e         // esat
      _L20) e (si_            _!e         // esi
      _L20) e (smu_           _!e         // esmu
+     _L20) e (sot_           _!e         // esot
        _m) e (dūzL04_          e
        _m) e (kL04_            e         // Meka
         m) e (lL88             E
@@ -261,6 +262,7 @@
        _n) ere (tL04_          ERE       // Nereta
         n) e (rv               E
         p) e (rs               e
+       pr) e (cL52z            e
        pr) e (L71L05_          e         // prece
       _pr) e (t                e
        pr) etē (@              ete:
@@ -287,6 +289,7 @@
         t) ermo                eRmo
         t) e (st               e
        _t) e (v_               e         // tev
+       _t) e (vi_              e         // tevi
        _v) e (lgL04_           e         // Velga
        _v) e (ltL04_           e         // Velta
        _v) e (ntL04_           E         // Venta
@@ -337,7 +340,7 @@
 
     // wide ē --------------------------------------------------------------------------- //
                                      // widening consonants follow
-           ē (L25              E:
+           ē (L25<             E:
            ē (L25L25           E:
                                      // widening vowels follow
            ē (L25L24+          E:
@@ -372,6 +375,7 @@
        gr) ē (kL01_            E:
       izp) ē (t                e:
      _kāp) ē (c_               e:    // kāpēc
+       _l) ē (L83+             e:    // lēst
         m) ēne (L83            e:ne
         m) ē (rķ               e:
        _m) ē (s_               e:    // mēs
@@ -425,7 +429,6 @@
            ī               i:
 
 .group j
-//        A) j (@           :j
            j               j
         _) jebk            j%ebk     // stress on 2nd syllable
 
@@ -483,11 +486,6 @@
            o (L41+         o         // o indicators after
            o (JL41+        o
 
-     L30J) o               uo        // uo indicators before (facultative writing)
-      L30) o (+            uo
-           o (JL30         uo        // uo indicators after (facultative writing)
-           o (L30+         uo
-
            o               uo        // words with Latvian roots are more common
            o (_++          uo        // common ending for Latvian words
            o (L40          uo        // usually uo before soft (Latvian) consonants
@@ -541,6 +539,7 @@ L46L45L45) o (<<           uo
     arist) o               o
      ater) o               o:
      _aut) o (<            o
+     bals) o               uo
     bandr) o               o:
      _bar) o (L12_         uo
       bar) o (L65          uo        // barot
@@ -575,12 +574,12 @@ L46L45L45) o (<<           uo
     dinam) o               o:
      disp) o               o
        eg) o               o>
+      eir) o               o
       _ek) o               o
       eks) o               o
     ekspl) o               o
     ekvin) o               o
    elektr) o               o:
-       ez) o               o>
     energ) o               o:
     enL87) o               o
       ent) o (@            o
@@ -588,6 +587,7 @@ L46L45L45) o (<<           uo
       epr) o               o
       ept) o (@            o
       eur) o               o
+       ez) o               o>
       gen) o               o
       get) o               o
      glik) o               o
@@ -692,6 +692,7 @@ L46L45L45) o (<<           uo
      pomp) o               o:
       por) o               o
        pr) o (L90          o
+      _pr) o               o
      prop) o               o
     pseid) o               o
      raps) o               o:
@@ -720,6 +721,7 @@ L46L45L45) o (<<           uo
       sol) o               o:
       son) o               o
      spid) o               o
+       sp) o (L77++        uo
     sport) o               uo
      sten) o (L65          uo
     _sten) o               o
@@ -763,6 +765,8 @@ L46L45L45) o (<<           uo
        gl) ob              ob
         g) ob (e           ob
         k) ob              ob
+        l) ob (ē           ob
+        l) oboto (m        oboto
         m) ob              ob
         _) ob (amL01_      ob        // Obama
            ob (je          ob
@@ -806,6 +810,7 @@ L46L45L45) o (<<           uo
         m) od (u           od
         m) od (uš          uod
         n) od (L05_        o:d       // ..node
+        n) od              uod
         _) od (ek          od
            od (erL51       uod
         _) od (esL04_      od        // Odesa
@@ -878,6 +883,7 @@ L46L45L45) o (<<           uo
        _d) ok (L01_        ok        // doks
         d) ok (u           ok
       fil) okok            okok
+       gl) ok (sīn         ok
         g) ok              ok
        _k) okako (lL04_    okako:    // kokakola
         k) ok (L01         uo`k      // koks with short uo, because of ambiguity
@@ -1063,6 +1069,7 @@ L46L45L45) o (<<           uo
        pr) om              om
       _pr) om (_           uom       // prom
        pt) om (L01_        o:m       // ..ptoms
+       pt) om              o>m
        rd) om (L01_        o:m
         r) om (b           om
        _r) om (L04_        uom       // Roma
@@ -1217,6 +1224,7 @@ L46L45L45) o (<<           uo
         m) op              op
       net) op (_           uop
         n) op (er+++       uop
+        n) opo (st         uopuo
            op (c           op
            op (en          op
         _) op (erL04_++++  o:p       // opera
@@ -1320,6 +1328,7 @@ L46L45L45) o (<<           uo
            or (ien         oR
         _) origo (_        oRigo:    // Origo
            or (iģ          oR
+           or (kā          oR
            or (ķest        oR
            or (L01_+       o:R       // ..ors
         _) or (L02_+       uoR       // ore
@@ -1387,6 +1396,7 @@ L46L45L45) o (<<           uo
         p) os (tC@         os
         p) os (te          os
         p) os (ter         os
+        p) os (tind        os
         p) os (tī          uos
         p) os (tL01        uos
         p) os (tL55m       os
@@ -1402,17 +1412,22 @@ L46L45L45) o (<<           uo
         š) os (ej          os
        _t) os (te          os
        _t) os (tL01_       os        // tosts
+        v) osto (k         osto
        _z) os (L06_        uos       // zoss
 
 .group ot
+
        an) ot              ot
+      ast) ot              uot
        av) ot              uot
      _bal) ot (ē           ot
+    bankr) ot              ot
         b) ot (ān          ot
         d) ot (āc          ot
         d) ot (ē           ot
        er) ot (iL67        o:t
        er) ot (is          o:t
+      _er) ot (L01_        ot        // Erots
       _ez) ot              ot
         f) oto (_          oto:
         g) ot (ik          o:t
@@ -1423,6 +1438,7 @@ L46L45L45) o (<<           uo
         k) ot (an          ot
         k) ot (L51         ot
         k) ot (l           ot
+    _L10pr) ot (L13_       uot       // (ne|sa..)prot(i|u)
       lin) ot (ip          ot
        _l) ot (iņL04       ot        // Lotiņa
         l) ot (L05_        ot        // ..lote
@@ -1433,22 +1449,25 @@ L46L45L45) o (<<           uo
      nepr) ot              uot
        _n) ot (L04_        ot        // nota
         n) ot (L50r        ot
+           ot (āj          uot
            ot (k           ot
+        m) otō             oto:
         _) oto (C<         oto
       pil) ot              ot
        _p) ot (ašL04       ot        // potaša
         p) ot (eL68        ot
+       pr) ot (am          uot
        pr) ot (e           ot
        pr) ot (o           ot
        pr) oto (š          uotuo     // protoš..
        pr) otot (ip        otot
        pr) oto (t          uotuo     // ..protot
-       pr) ot (u_          uot
   publisk) ot              uot
         r) ot (āc          ot
         r) ot (b           ot
         r) ot (er          ot
         r) ot (ē           ot
+        r) ot (kaL77       o`t
       sab) ot              ot
        sk) ot (L01_        ot
        sl) ot (L04_        uot       // slota
@@ -1566,6 +1585,7 @@ L46L45L45) o (<<           uo
            v               v
                                        // unstressed if starts as vis.. prefix
                                        // except for words where vis.. is part of root
+           vecvec          vetsvets      
         _) vi (sād         v'i         // visād
         _) vi (sēd         v'i         // visēd
         _) vi (skijL01_    v'i         // viskijs

dictsource/sd_list

@@ -1,170 +1,194 @@
-sindhi
-// This file is UTF8 encoded 
-// Letter Names, Numbers and  exceptional pronounciation of  Urdu
-
-
-// The Letter names
-ا	alIf
-ب	be:
-پ	pe:
-ت	te:
-ٹ	te:
-ث	se:
-ج	Ji:m
-چ	ce:
-ح	bar.i:||
-خ	xe:
-د	da:l
-ڈ	da:l
-ذ	za:l
-ر	Re:
-ڑ	r.e:
-ز	ze:
-ژ	Ze:
-س	si:n
-ش	Si:n
-ص	sUa:d
-ض	zUa:d
-ط	tOe:
-ظ	zOe:
-ع	aIn
-غ	QaIn
-ف	fe:
-ق	qa:f
-ک	ka:f
-گ	ga:f
-ل	la:m
-م	mi:m
- ن	nu:n
-_و	va:O
-ہ	he:
-ں	ghUnna
-ﮨ	cho:ti:||he:
-ھ	dO||caSmi:||he:
-ء	hamzah
-ی	c#o:ti:||je:
-ے	bar.i:||je:
-ئ	h'amza
-ۂ	he:h'amza
-_ؤ	va:Oh'amza
-
-َ	z@b@R
-ِ	ze:R
-ُ	pe:S
-ٰ	,alIfmaqs'u:r,a
-ّ	t,aSd'i:d
-ً	d'o:zab'ar
-
-
-// The numbers
-_0	bu're
-_1 	hi'ku
-_2 	ba:
-_3 	te
-_4 	ca:r
-_5 	pan'ja:
-_6 	ch'aha:
-_7 	sa'ta:
-_8 	ath'a:
-_9 	na'va:
-_10 	da'ha:
-_11 	jara:ha
-_12 	bara:ha
-_13 	tera:ha
-_14 	cOda:ha
-_15 	pa'dra:ha
-_16 	sora:ha
-_17 	satira:ha
-_18 	ari'rahan
-_19 	una'via
-_20 	viha:
-_21 	ek'kie
-_22 	bavi'a
-_23 	tevi'a
-_24 	cO'via
-_25 	panc'via
-_26 	cha'via
-_27 	sat'avia
-_28 	at:avia
-_29 	una:tie
-_30 	ti'ha
-_31 	ek'atiha
-_32 	battiha
-_33 	tettiha
-_34 	cO'tiha
-_35 	panc:tiha
-_36 	chat'tiha
-_37 	sat'tatiha
-_38 	att:atiha
-_39 	unuta:vie
-_40 	ca'li:
-_41 	ek'eta:rie
-_42 	ba:ite'rie
-_43 	te'ta:rie
-_44 	cOt'erie
-_45 	panci'terie
-_46 	cai'terieh
-_47 	sati'terie
-_48 	ati'terie
-_49 	urur'ja
-_50 	pan'ja
-_51 	ekvan'ja
-_52 	bavan'ja
-_53 	tevan'ja
-_54 	cOvanja
-_55 	pa:van'ja
-_56 	ca'vaja:
-_57 	sat:ur'ja
-_58 	at:ur'ja
-_59 	una:hat:i
-_60 	sa:thi
-_61 	eka:thi
-_62 	ba:hathi
-_63 	teha:thi
-_64 	cO'hathi
-_65 	panja'hathi
-_66 	ca:hathi
-_67 	sat'tathi
-_68 	at'thati
-_69     un'ha't:har
-_70 	sata:ri
-_71 	ek'hat:ar
-_72 	ba'hat'ar
-_73 	te'hat'ar
-_74 	cOhat'ar
-_75 	pan'jat'ar
-_76 	cha'hat'ar
-_77 	sat'ta'tar
-_78 	at:hat'ar
-_79 	un'a:si
-_80 	a:si
-_81 	ek'ea'si
-_82 	bea'si
-_83 	tea'si
-_84 	cO'rasi
-_85 	panc'a:si
-_86 	chhajasi
-_87 	sat'tea:si
-_88 	at:ha:si
-_89 	ur'ra:nave
-_90 	nave
-_91 	ek'ka'navi
-_92 	bea'navi
-_93 	tea'navi
-_94 	cOra:navi
-_95 	panca:navi
-_96 	cha:navi
-_97 	sat'ta:navi
-_98 	at:tha:navi
-_99 	nava:navi
-_100 	sO
-
-_0M1	haz'a:r
-_0M2	l'a:kh
-_0M3	kaR'o:R
-_0M4	'aRab
-_0M5	'k#aRab
-
-_dpt	_da:he_
-
-
+// This file is UTF8 encoded 
+// Letter Names, Numbers and  exceptional pronounciation of  Sindhi
+
+// The Letter names
+ا	alIf
+ب	be:
+ٻ	bE
+ڀ	b#e:
+پ	pe:
+ت	te:
+ٿ	t#e:
+ٽ	t.e:
+ٺ	t.#e:
+ث	se:
+ج	Ji:m
+ڄ	Je:
+جھ	J#e:
+ڃ	n^e:
+چ	ce:
+ڇ	c#e:
+ح	vad.i:||he:
+خ	xe:
+د	da:l
+ڌ	d#a:l
+ڊ	d.a:l
+ڏ	d.e:
+ڍ	d.#a:l
+ذ	za:l
+ر	Re:
+ڙ	r.e:
+ز	ze:
+ژ	Ze:
+س	si:n
+ش	Si:n
+ص	sUa:d
+ض	zUa:d
+ط	tOe:
+ظ	zOe:
+ع	aIn
+غ	QaIn
+ف	fe:
+ڦ	p#e:
+ڪ	ka:f
+ق	qa:f
+ک	k#e:
+گ	ga:f
+گھ	g#e:
+ڳ	ge:
+ڱ	N-e:
+ل	la:m
+م	mi:m
+ن	nu:n
+ڻ	n.u:n.
+و	va:O
+ه	he:
+ہ	Urdu:||he:
+ھ	dO||caSmi:||he:
+ء	hamzah
+ي	je:
+ئ	h'amza||je:
+ۂ	he:h'amza
+ؤ	va:Oh'amza
+
+َ	z@b@R
+ِ	ze:R
+ُ	pe:S
+ٰ	,alIfmaqs'u:r,a
+ّ	t,aSd'i:d
+ً	d'o:zab'ar
+ٗ	Ubto:pe:S
+
+
+// The numbers
+_0	bUr.i:
+_1	hIkU
+_2 	ba:
+_3	t.e
+_4	ca:R
+_5	panJa
+_6	c#aha
+_7	sat:a
+_8	at.#a
+_9	nava
+_10	d.aha
+_11	jaraha
+_12	baraha
+_13	te:raha
+_14	cOd.aha
+_15	pandraha
+_16	soraha
+_17	satraha
+_18	arr.aha
+_19	Un.vi:ha
+_20	vi:ha
+_21	Ikvi:ha
+_22	ba:vi:ha
+_23	t.e:vi:ha
+_24	co:vi:ha
+_25	panJvi:ha
+_26	c#avi:ha
+_27	sata:vi:ha
+_28	at.#a:vi:ha
+_29	Un.at.i:ha
+_30	t.i:ha
+_31	Ikat.i:ha
+_32	bat.i:ha
+_33	t.e:t.i:ha
+_34	co:t.i:ha
+_35	panJati:ha
+_36	c#at.i:ha
+_37	satati:ha
+_38	at.#ati:ha
+_39	Un.ta:li:ha
+_40	ca:li:ha
+_41	Ikta:li:ha
+_42	ba:Ita:li:ha
+_43	t.e:ta:li:ha
+_44	co:Ita:li:ha
+_45	panJe:ta:li:ha
+_46	c#a:Ita:li:ha
+_47	sate:ta:li:ha
+_48	at.#e:ta:li:ha
+_49	Un.vanJa:h
+_50	panJa:h
+_51	IkvanJa:h
+_52	ba:vanJa:h
+_53	t.e:vanJa:h
+_54	co:vanJa:h
+_55	panJvanJa:h
+_56	c#a:vanJa:h
+_57	satvanJa:h
+_58	at.#vanJa:h
+_59	Un.ahat.#I
+_60	sat.#I
+_61	Ikahat.#I
+_62	bahat.#I
+_63	t.e:hat.#I
+_64	co:hat.#I
+_65	panJahat.#I
+_66	c#a:hat.#I
+_67	satahat.#I
+_68	at.#ahat.#I
+_69	Un.hat:aR
+_70	sat:aR
+_71	Ikahat:aR
+_72	ba:hataR
+_73	t.e:hataR
+_74	co:hataR
+_75	panJataR
+_76	c#a:hataR
+_77	satat:aR
+_78	at.#ahataR
+_79	Un.Ia:si:
+_80	assi:
+_81	IkIa:si:
+_82	bIa:si:
+_83	t.e:a:si
+_84	cOo:ra:si:
+_85	panJa:si:
+_86	c#aha:si:
+_87	satIa:si:
+_88	at.#Ia:si:
+_89	Un.Ia:nave:
+_90	navve:
+_91	IkIanave:
+_92	bIa:nave:
+_93	t.Ia:navE:
+_94	co:Ra:nave:
+_95	panJa:nave:
+_96	c#aha:nave:
+_97	satIa:nave:
+_98	at.#Ia:nave:
+_99	nava:nave:
+
+_0C	s'aU
+_0M1	haz'a:r
+_0M2	l'a:kh
+_0M3	kaR'o:R
+_0M4	'aRab
+_0M5	'k#aRab
+
+_dpt	_d.aha:i:_
+
+// word exceptions
+
+آھي	a:he:	$u
+آهن	a:hIn	$u
+هن	hIna
+۾	me~	$u
+۽	ae~	$brk
+آهي	a:he:	$u
+هڪ	HIk
+ٿي	t#i:	$u
+ئي	i:

dictsource/tr_list

@@ -30,7 +30,7 @@ t	tE
 u	u
 ü	y
 v	vE
-w	tS'ift||vE
+w	d'ab@LjU
 x	iks
 y	jE
 z	zE

dictsource/ur_list

@@ -1,4 +1,4 @@
-// This ur_list file is prepared  by Him Prasad Gautam (Kathmandu, Nepal)
+// This ur_list file is prepared  by Him Prasad Gautam (Kathmandu, Nepal) and Ejaz Shah (Pakistan)
 // This file is UTF8 encoded 
 // Letter Names, Numbers and  exceptional pronounciation of  Urdu
 
@@ -55,6 +55,8 @@ _ؤ	va:Oh'amza
 ّ	t,aSd'i:d
 ً	d'o:zab'aR
 ْ	sUk'u:n
+ٗ	Ult.a:||pe:S
+ٓ	mada
 
 
 // The numbers
@@ -190,8 +192,11 @@ _dpt	_ISaRI'Ia_
 جن	J'In	$u
 جنھیں	J,InH'e:n	$u
 انہوں	,UnH'o:n	$u
+انھوں	UnHo:n	$u
 انہیں	,UnH'e:n	$u
 انھیں	,UnH'e:n	$u
+انھی	UnHi	$u
+انہی	UnHi:	$u
 
 // unstressed prepositions
 کا	$u
@@ -199,7 +204,7 @@ _dpt	_ISaRI'Ia_
 کے  	$u
 سے 	$u
 
-اور 	OR	$u $brk
+اور 	aUR	$u $brk
 کہ	k'e:H	$u
 نے	$u
 
@@ -329,7 +334,7 @@ _dpt	_ISaRI'Ia_
 حفاظت	H,Ifa:z.'at
 ہمت	HIm'at
 سمیت	sam'e:t
-اگرچہ	,agaRc'e:H
+اگرچہ	,agaRc'e:H	$pause
 بلکہ	b,alk'e:H
 حصہ	H'Is.,a
 حصے	H'Is.e:
@@ -677,14 +682,13 @@ _dpt	_ISaRI'Ia_
 کلک	klIk
 فکسنگ	f,Iks'Ing
 مانند	ma:n'Ind	$1
-اتار	Ut'a:R
 آرکائیو	,a:Rk'a:jv
 لنک	lInk
 فہرست	fEHRIst
 کیلئے	ke:li:;e:
-اگر	agaR	$u
+اگر	agaR	$pause
 ضرورت	zaRu:Rat
-مگر	magaR	$u
+مگر	magaR	$pause
 دیگر	digaR
 اہم	EHEm
 فیصد	fis.ad
@@ -718,4 +722,273 @@ _dpt	_ISaRI'Ia_
 نصرت	nUs.Rat
 خاتون	xa:tu:n
 برتری	baRtaRi
-برتر	baRtaR
+برتر	baRtaR
+سن	sUn
+انھتر	UnHataR
+انیس	Uni:s
+ورلڈ	vaRld.
+فرق	faRq
+اوپر	u:paR	$1
+اوپری	u:pRi:
+دودھ	du:d#
+اورئینٹل	o:Ri;e:nt.al
+انکار	Inka:R
+انکاری	Inka:Ri
+انیک	ane:k
+اونگ	u:ng
+اوک	u:k
+اویس	ave:s
+اویسی	ave:si
+اٹوٹ	at.u:t.
+اٹھ	Ut.#
+اٹھان	Ut.#a:n
+اک	Ik
+اکا	Ika:
+دکا	dUka:
+اکائی	Ik'a:i
+اکبر	akbaR	$1
+وکی	vIki
+اکرم	akRam	$1
+اکزیکٹیو	Ikze:kIt.Iv
+اکتا	Ukta:
+اکتایا	Ukta:ja:
+اکتاہٹ	Ukta:Hat.
+اکلو	Iklo:
+اکمل	akmal	$1
+مملکت	mamlIkat
+قربت	qURbat	$1
+مٹا	mIt.a:
+ٹیوب	t.ju:b
+سیہون	sEHvan
+عرس	URs
+اکھاڑ	Uk#a:r.
+اکے	Ike:
+آمیزش	a:me:zIS
+واحد	va:HId
+آنت	a:nt
+آنچل	a:ncal
+حجاب	HIJa:b
+حلقہ	Halqa
+لحاظ	lIHa:z.
+کیچڑ	kicar.
+گردش	gaRdIS
+مروت	mURavvat
+مزاج	mIza:J
+غربت	QURbat	$1
+آویزش	a:ve:zIS
+پہر	pEHER
+آہ	a:H
+جگر	JIgaR
+دلدوز	dIldo:z	$1
+رحمت	REHmat
+ابجد	abJad
+(رسم الخط)	RasmUlxat
+ابجدی	abJadi
+حروف	HURu:f
+ابر	abR
+ابراہیم	IbRa:Hi:m
+ابراہیمی	IbRa:Hi:mi
+ابرق	abRaq
+ابرنا	UbaRna:
+ابرو	abRu:
+کنول	kV~val	$2
+ابل	Ubal
+ابلا	Ubla:
+ابلی	Ubli:
+ابلیس	Ibli:s
+الغرض	alQaRaz
+الوقت	Ulvaqt
+دہر	dEHER
+ظلم	z.Ulm
+ابو	abu:
+طالب	ta:lIb
+اپج	UpaJ
+اپچار	Upca:R
+اپدیش	Upde:S
+اپواس	Upva:s
+اپکار	Upka:R
+سرسری	saRsaRi	$1
+متخالف	mUtaxa:lIf
+اتصالی	,ItIs.a:li
+اتمام	ItIma:m
+حجت	HUJat
+اتھل	Ut#al
+پتھل	pUt#al
+اجازت	IJa:zat	$2
+اجرا	IJRa:
+اجرت	IJRa:
+اجگر	aJgaR
+اجلا	UJla:
+اجلی	UJli:
+اجلے	UJle:
+اجماع	IJma:
+اجمل	aJmal
+اجنبی	aJnabi
+اجنٹ	IJant.
+اجنڈا	IJand.a:
+اچک	Ucak
+اچکا	Ucka:
+اجنب	aJnab
+اچھوت	ac#u:t
+اچھوتا	ac#u:ta:
+اچھوتی	ac#u:ti
+اچھوتے	ac#u:te:
+پرسی	pURsi:
+بالغیب	bIlQEb	$2
+نگار	nIga:R
+کاغذ	ka:QaD
+اختری	axtaRi:
+اخذ	axaD
+جرات	JURat
+اخوت	Uxu:vat
+فہم	fEHEm
+جرمنی	JaRmani
+خدمت	xIdmat
+شریعت	SaRi;at
+ادت	Idat
+ادراک	IdRa:k
+تردد	taRaddUd
+خدمات	xIdma:t
+ادریس	IdRi:s
+ادب	adab
+ادغام	IdQa:m
+ادلال	Idla:l
+موا	mu:a:
+ادھر	Id#aR
+(ادھر ادھر)	Id#aR||Ud#aR
+ادھم	Ud#am
+اذالہ	IDa:la
+ارتھ	aRt#
+معلیٰ	mUalla:
+ارسال	IRsa:l
+ملیں	mIle:n
+ارشاد	IRSa:d
+ارض	aRz
+سلطنت	saltanat
+استاد	Usta:d
+استادی	Usta:di
+استر	staR
+بخدا	baxUda:
+بندش	bandIS
+خرد	xURd
+خردار	xURda:R
+خردہ	xURda
+خرشید	xURSi:d
+تبرک	tabaRUk
+خرم	xURRam
+خزاں	xIza:n
+پروردہ	paRvaRda
+خشک	xUSk
+خصوص	xUs.u:s.
+زمرد	zamaRUd
+جبلت	JIbIll'at	$2
+بالآخر	bIla:xIR
+شعاع	SUa:
+حصار	HIs.a:R
+شدگی	SUdgi:
+بالاتفاق	bIlItIfa:q
+بالاجماع	bIlIJma:
+بالاجمال	bIlIJma:l
+بالاشتراک	bIlIStIRa:k
+بالاعلان	bIle:la:n
+بالجبر	bIlJabR
+بالعوض	bIlIvaz
+بالغرض	bIlQaRaz
+مرکزی	maRkazi
+بالفرض	bIlfaRz
+حکمت	HIkmat
+بالفعل	bIlfe:l
+بالمثل	bIlmIsl
+بالمختصر	bIlmUxtas.aR
+بالمشافہ	bIlmUSa:fa
+بالمعنی	bIlma:ni
+بالمقابل	bIlmUqa:bIl
+بالواسطہ	bIlva:sta
+نیند	ni:nd
+باوجودیکہ	ba:vUJu:de:ke:
+کتا	kUt:a:	$2
+دکھاؤ	dIk#a:o:
+کتے	kUt:e:	$2
+کتوں	kUt:o:n
+بپھر	bIp#aR
+شکن	SIkan
+بتکرار	batakRa:R
+بچھ	bIc#
+منجمد	mUnJamId
+بحر	bEHER
+بخدمت	baxIdmat
+منکشف	mUnkaSIf
+بخش	baxS
+بخشش	baxSIS
+جدت	JIdat
+بخوشی	baxUSi:
+بخیر	baxE:R
+بچپن	bacpan
+بچپنا	bacpana:
+قطب	qUtUb
+ذہن	zEHEn
+سلوک	sUlu:k
+انڈین	Ind.i;an
+متنبہ	mUtanabe:
+قیمتی	qi:mti:
+بدک	bIdak
+بذریعہ	baDaRi:;a:
+برا	bURa:
+نسبتی	nIsbati
+برتن	baRtan
+برج	bURJ
+برد	bURd
+بردہ	bURda
+رغبت	RaQbat
+برضا	baRaza:
+علاج	Ila:J
+الفت	Ulfat
+برگزیدہ	baRgUzi:da
+برہمن	bRaHman
+الذمہ	UzzIma
+صحبت	s.OHbat
+برے	bURe:
+بز	bUz
+بزم	bazm
+بستر	bIstaR
+مرگ	maRg
+بسکٹ	bIskIt.
+(بسم اللہ)	bIsmIl:a:H
+بسم	bIsm
+بسنت	basant
+بضد	bazId
+بعید	bai:d
+بغض	bUQz
+بقید	baqE:d
+(بلا واسطہ)	bIla:||va:sta
+بلال	bIla:l
+بلبل	bUlbUl
+بلبلے	bUlbUle:
+بلک	bIlak
+بلیک	ble:k
+مارکٹ	ma:RkIt.
+(بن بلائے)	bIn||bUla:e:
+(بن مول)	bIn||mo:l
+بنت	bInt
+سنور	sV~vaR
+بو	bu:	$only
+بورڈنگ	bo:Rd.Ing
+بولنگ	bo:lIng
+امرت	amRIt
+(ان بجھا)	anbUJ#a:
+(ان بن)	anban
+(ان پڑھ)	anpar.h
+(ان گنت)	an||gInat
+قدرت	qUdRat
+انسلٹ	Insalt.
+انسیت	Unsi;at
+اللہ	al:a:H
+انصاف	Ins.a:f
+انصافی	Ins.a:fi
+تپش	tapIS
+انثیت	Unsi;at
+انجم	anJUm
+انجمن	anJUman
+الجھن	UlJ#an
+الحمد	alHamd
+التفات	IltIfa:t

dictsource/ur_rules

@@ -1,5 +1,5 @@
 
-// This ur_rules file is written by Him Prasad Gautam (Kathmandu, Nepal).
+// This ur_rules file is written by Him Prasad Gautam (Kathmandu, Nepal) and Ejaz Shah (Pakistan).
 // This file is UTF8 encoded 
 //  phoneme rules for Urdu
 // Letter group - l01: Vowels
@@ -17,16 +17,18 @@
 ٩	9
 ي	ی
 یٰ	ا
+ٰی	ا
+آ	آ
 
 
 // vowel signs, alif acts as consonants  at start of word if these follow
 .L01  َ ِ ُ و ی ع ا ے 
 
 // Consonants
-.L02  ب پ ت ٹ ث ج چ ح خ د ڈ ذ ر ڑ ز ژ س ش ص ض ط ظ غ ف ق ک گ ل م ن ہ ئ
+.L02  ب بھ پ پھ ت تھ ٹ ٹھ ث ج جھ چ چھ ح خ د دھ ڈ ڈھ ذ ر ڑ ز ژ س ش ص ض ط ظ غ ف ق ک کھ گ گھ ل م ن ہ ئ
 
 // Letter group - L03:  Most of the letters
-.L03  ا ب پ ت ٹ ث ج چ ح خ د ڈ ذ ر ڑ ز ژ س ش ص ض ط ظ ع غ ف ق ک گ ل م ن و ہ ئ ی ے
+.L03  ا ب بھ پ پھ ت تھ ٹ ٹھ ث ج جھ چ چھ ح خ د دھ ڈ ڈھ ذ ر ڑ ز ژ س ش ص ض ط ظ ع غ ف ق ک کھ گ گھ ل م ن و ہ ئ ی ے ٰ
 
 // VOWELS
 
@@ -34,7 +36,7 @@
 	َ	a
 	َی 	E
 	َے	E
-	َو 	O
+	َو 	aU
 	َع	a:
 
 .group ِ
@@ -48,6 +50,9 @@
 	ُو	u:
 	ُع	O
 
+.group ٗ
+	ٗ	u:
+
 .group ھ
 	ھ	HV 
 	ھ (_	H:
@@ -80,7 +85,7 @@ _) ای	e:
 	افتتاح	,ItIt'a:H
 	اسلام	,Isl'a:m
 	_) انسان	Ins'a:n
-	_) انت (L02	'Int,a
+	_) انت (L02L01	'Int,a
 	_) امید	Um'i:d
 	اً(_S2	'an
 	_) امکان	,Imk'a:n
@@ -123,12 +128,65 @@ _) ای	e:
 	_) است (L02	,IstI
 	_) ارت (L02	,IRtI
 	_) ادار	Ida:R
+	_)انک(L02L01	InkI
+	_)انہتر	UnHataR
+	_)او(ا	av
+	_)انہ(L02L01	InHI
+	_)اٹھتال	at.#ta:l
+	_)اٹھتر	at.#at:aR
+	_)اٹھتیس	at.#ti:s
+	_)اٹھلا	It.#la:
+	_)اٹھ(و	Ut.#
+	_)اکت(L02L01	IktI
+	_)اکسا	Uksa:
+	_)اکٹھ(L01	Ikat.t.#
+	_)اکہر	IkEHR
+	_)اکیل	ake:l
+	_)ائ	aI
+	_)ابال	Uba:l
+	_)ابوال(P5	abUl
+	L02)او(L02	a:vV
+	_)ابھار	Ub#a:R
+	_)اتار	Uta:R
+	_)اترو(L01	UtaRv
+	_)ابکائ	Ubka:
+	_)اجت(L02L01	IJtI
+	_)اجا(ر	IJa:
+	_)اجال	UJa:l
+	_)اجاڑ	UJa:r.
+	_)اچھال	Uc#a:l
+	_)اداس	Uda:s
+	_)ادھار	Ud#a:R
+	_)ادھڑ(L01	Ud#r.
+	_)ادھڑ	Ud#ar.
+	_)ادھیڑ	Ud#e:r.
+	_)ارا(د	IRa:
+	_)اونچ	u:nc
+	اں	a~
+	_)انج(L02L01	InJI
+	_)امت(L02L01	ImtI
+	_)انح(L02L01	InHI
+	_)انخ(L02L01	InxI
+	_)انس(پ	Ins
+	_)انس(L02L01	InsI
+	_)انس(ٹ	Ins
+	_)انسٹھ	Unsat.#
+	_)انشا	InSa:
+	_)انش(L02L01	InSI
+	_)انص(L02L01	Ins.I
+	_)انض(L02L01	InzI
+	_)انق(L02L01	InqI
+	_)الت(L02L01	IltI
+	_)انف(L02L01	InfI
+	_)ال(L03L03L03P2	al
 
 .group آ
 	آ	a:
 	آتش	'a:tIS
 	آہست	a:H'Ist
 	آئ (L02	'a:I
+	آمیز	a:me:z
+	آویز	a:ve:z
 
 
 .group ع
@@ -141,6 +199,8 @@ _) ای	e:
 	_) عمر	'UmR
 	L02) عی (_	'ai:
 	علیحد	'alEHd
+	عُ	'U
+	عِ	I
 
 .group و
 	و	o:
@@ -152,8 +212,11 @@ _) ای	e:
 	وّ	vv
 	L02L02L02) و (L02	u:
 	L02) ونکہ (_	u:nk'e:H
-	L02L01) و (L02	vV
-	L01) و (L01	v
+	_L02)و(L01	av
+	L01)و(L01	v
+	وؤ	u:o:
+	L03L03L03L03)وں(S2	o:n
+	_L02L01)و(L02_	vV
 
 .group ی
 	ی	e:
@@ -162,12 +225,17 @@ _) ای	e:
 	_) ی	jV
 	یع(_	i:
 	@) ی	i:
-	@) یں (_	e:n
+	@) یں (_	e~
 	یّ	jj
 	@) یت (_S2	i:;'at
 	_) یو	jO
 	ی (و	e:
 	L02)یوں	i;o:n
+	L03)یکشن	e:kSan
+	L03L03L03L03)یں(S2	e~
+	یں	e~
+	L03)یک(ٹ	e:k
+	L03L01)یا(_S2	ja:
 
 .group ے		// yeh barree
 	ے	e:
@@ -180,7 +248,7 @@ _) ای	e:
 	ئ
 
 .group ؤ
-	ؤ	aU   // ??
+	ؤ	o:
 
 .group أ
 	أ	// ??
@@ -220,6 +288,12 @@ _) ای	e:
 	بوڑھ	b'u:r.h
 	_) بے (P2	b'e:
 	بحران	b,OHR'a:n
+	_)بپھر(L01	bIp#R
+	_)بچھ	bIc#
+	_)بخار	bUxa:R
+	_)برائی	bURa:i
+	_)بوا(ئ	boa:
+	_)بزدل	bUzdIl
 
 
 .group پ
@@ -237,6 +311,7 @@ _) ای	e:
 	پیچھ	p'i:c#
 	پھونک	p#'u:nk
 	_) پور (L01	p'u:R
+	_)پچھل(L01	pIc#l
 
 .group ت
 	ت	tV
@@ -254,10 +329,11 @@ _) ای	e:
 	تجرب	taJ'aRb
 	_) تیار	t,ai:;'a:R
 	_) تیسر	t'i:sR
-	L03L03) تا (_S2	ta:
+	L03L03)تا(_S2	ta:
 	L03L03) تی (_S2	ti:
 	L03L03) تے (_S2	te:
 	_) توا	t'ava:
+	تہ(_	ta
 
 .group ٹ
 	ٹ	t.V
@@ -295,6 +371,7 @@ _) ای	e:
 	_) جمل	J'Uml
 	_) جہ (L02	J'EH
 	_) جدا	J'Uda:
+	_)جذب	JaDb
 
 .group چ
 	چ	cV
@@ -320,6 +397,7 @@ _) ای	e:
 
 	حّ	HHV
 	حّ (L01	HH
+	_)حیا	Haja:
 
 .group خ
 	خ	xV
@@ -331,6 +409,8 @@ _) ای	e:
 	_) خود	x'Ud
 	_) خواہش	xv'a:HIS
 	_) خیا	x'aja:
+	_)خوا	xva:
+	_)خدا	xUda:
 
 
 .group د
@@ -353,6 +433,7 @@ _) ای	e:
 	_) دوسر	d'u:sR
 	_) دین	de:n
 	_) دیت	de:t
+	دہ(_	da
 
 .group ڈ
 	ڈ	d.V
@@ -366,11 +447,9 @@ _) ای	e:
 	ڈرائیور	d.R'a:jvaR
 
 .group ذ
-	ذ	DV
-	ذ (L01	D
-	ذّ	DDV
-	ذّ (L01	DD
-	_) ذہ (L02	D'EH
+	ذ	zV
+	ذ (L01	z
+	_) ذہ (L02	z'EH
 
 .group ر
 	ر	RV
@@ -446,6 +525,7 @@ _) ای	e:
 	_) شخص	S'axs.
 	_) شہ (L02	S'EH
 	شرمند	S,aRm'Ind
+	_)شمار	SUma:R
 
 .group ص
 	ص	s.V
@@ -529,12 +609,13 @@ _) ای	e:
 	_) گہ (L02	g'EH
 	گزارش	gUz'a:RIS
 	_) گیت	g'i:t
+	_)گزار	gUza:R
 
 .group ل
 	ل	lV
 	ل (L01	l
-	لّ	llV
-	لّ (L01	ll
+	لّ	l:V
+	لّ (L01	l:
 	لکھ	l'Ik#
 	_) لہ (L02	l'EH
 	_) لیت	le:t
@@ -596,12 +677,12 @@ _) ای	e:
 	_) نمٹ	nIm'at.
 	_) نمٹ (L01	nImt.
 	@) ن (پ	n
-	@) نچل	n'Icl
+	_) نچل	n'Icl
 	نشان	nIS'a:n
 	@) ن (ڈ	n
 	_) نظر	naz.'aR
 	_) نظام	nIz.'a:m
-	_) نہ (L02	n'H
+	_) نہ (L02	nEHE
 	L01) نٹ	n't.
 	نھ	nH
 	_) نکت	n'Ukt
@@ -626,8 +707,8 @@ _) ای	e:
 	ہ	H
 	_) ہ (L01	H
 	_) ہ (ند	HI
-	ہّ	HHV
-	ہّ (L01	HH
+	ہّ	H:V
+	ہّ (L01	H:
 
 	ہ (_	a
 	_) ہلن	H'Iln
@@ -636,24 +717,25 @@ _) ای	e:
 	_) ہم (L02L02L02P2	H'am
 
 .group ّ
-	ّ
+	ّ	:
 
 .group ً
-	ً
+	ً	an
 
 
 // Common symbols
 
 .group
 	$	do:l@R
-	,	sVkat.Vh	// 	Sakata سکتہ
-		!	f'iJa:ja	//Fijaya	فجائیہ
-	;	w'ak't'a	//Waqfa	وقفہ
+	,	sakta	// 	Sakta سکتہ
+	!	f'iJa:ja	//Fijaya 	فجائیہ
+	;	vakfa	//Waqfa 	وقفہ
 	:	r'a:b'tah//Rabtar 	رابطہ
 	?	saw'a:l'i:ja:n'is'a:nV	// Sawaliya Nishan	سوالیہ نشان
-	%	f'i:sVd	//Fisad	فیصد
+	%	f'i:sVd	//Fisad 	فیصد
+	۔	xatma	//Khatma 	ختمہ
+	؟	sava:li:;a:||nISa:n	//Sawalya Nishan 	سوالیہ نشان
 
-	
 
 .group 0xe0a4  // devanagari
 	0xe0a4  _^_NE  // switch to Nepali voice

docs/dictionary.md

@@ -55,9 +55,9 @@ The utility 'phonemes' are:
 | `=`    | put the primary stress on the preceding syllable |
 | `_:`   | short pause |
 | `_`    | a shorter pause |
-| `||`   | indicates a word boundary within a phoneme string |
-| `|`    | can be used to separate two adjacent characters, to prevent them from being considered as a multi-character phoneme mnemonic |
- 
+| <code>&#124;</code>  | indicates a word boundary within a phoneme string |
+| <code>&#124;&#124;</code>    | can be used to separate two adjacent characters, to prevent them from being considered as a multi-character phoneme mnemonic |
+
 It is not necessary to specify the stress of every syllable. Stress
 markers are only needed in order to change the effect of the language's
 default stress rule.

docs/languages.md

@@ -8,7 +8,7 @@ and dialects,
 [private-use extensions](https://raw.githubusercontent.com/espeak-ng/bcp47-data/master/bcp47-extensions)
 have been used.
 
-The supported languages are:
+The 100 supported languages and accents are:
 
 | Family Code | Identifier        | Language Family       | Language                    | Accent/Dialect         |
 |-------------|-------------------|-----------------------|-----------------------------|------------------------|
@@ -75,6 +75,7 @@ The supported languages are:
 | `art`       | `lfn`             | Constructed           | Lingua Franca Nova          |                        |
 | `bat`       | `lt`              | Baltic                | Lithuanian                  |                        |
 | `art`       | `jbo`             | Constructed           | Lojban                      |                        |
+| `poz`       | `mi`              | Malayo-Polynesian     | Māori                       |                        |
 | `zls`       | `mk`              | South Slavic          | Macedonian                  |                        |
 | `poz`       | `ms`              | Malayo-Polynesian     | Malay                       |                        |
 | `dra`       | `ml`              | Dravidian             | Malayalam                   |                        |

emscripten/demo.html

@@ -52,6 +52,12 @@
         display: block;
         margin-bottom: 1rem;
       }
+      #ipaarea {
+        width: 100%;
+        height: 8rem;
+        display: block;
+        margin-bottom: 1rem;         
+      }
       p.gh {
         color: #333;
         text-align: right;
@@ -100,6 +106,7 @@
     <h1>espeakng.js 1.49.1 Demo</h1>
     <form>
       <textarea id="texttospeak">Call me Ishmael. Some years ago --- never mind how long precisely --- having little or no money in my purse, and nothing particular to interest me on shore, I thought I would sail about a little and see the watery part of the world. It is a way I have of driving off the spleen and regulating the circulation.</textarea>
+      <textarea readonly="true" placeholder="IPA" id="ipaarea"></textarea>
       <div class="speecharg">
         <label for="pitch">Pitch</label><input id="pitch" type="range" value="50" min="0" max="100" /><button type="button" aria-label="Reset pitch" title="Reset pitch" onclick="resetPitch();">&#x21b6;</button>
       </div>
@@ -110,7 +117,7 @@
         <label for="voice">Voice</label><select id="voice"></select><button type="button" aria-label="Reset voice" title="Reset voice" onclick="resetVoice();">&#x21b6;</button>
       </div>
       <div class="bottom">
-        <button type="button" onmousedown="speak();">Speak</button>
+        <button type="button" onmousedown="speakAndIpa();">Speak</button>
         <button type="button" onmousedown="stop();">Stop</button>
       </div>
       <p class="gh"><a href="https://github.com/eeejay/espeak/tree/emscripten">Original Code for eSpeak: Eitan Isaacson</a></p>

emscripten/espeakng_glue.cpp

@@ -50,6 +50,35 @@ public:
     espeak_SetSynthCallback(NULL);
   }
 
+  int synth_ipa_(const char* aText, const char* virtualFileName) {
+
+  /* phoneme_mode
+    bit 1:   0=eSpeak's ascii phoneme names, 1= International Phonetic Alphabet (as UTF-8 characters).
+    bit 7:   use (bits 8-23) as a tie within multi-letter phonemes names
+    bits 8-23:  separator character, between phoneme names
+  */
+    
+    espeak_SetSynthCallback(NULL);
+        
+    int phoneme_options               = (1 << 1); // Use IPA
+    int use_custom_phoneme_separator  = (0 << 7);
+    int phonemes_separator            = ' '; // Use a default value
+
+    int phoneme_conf                  = phoneme_options | (phonemes_separator << 8);
+    
+    FILE* f_phonemes_out = fopen(virtualFileName,"wb");
+    if(!f_phonemes_out)
+      return -1;
+    
+    //espeak_ng_InitializeOutput(ENOUTPUT_MODE_SYNCHRONOUS, 0, NULL);
+    espeak_SetPhonemeTrace(phoneme_conf, f_phonemes_out);
+    espeak_Synth(aText, 0, 0, POS_CHARACTER, 0, 0, NULL, NULL);
+    espeak_SetPhonemeTrace(0, NULL);
+    fclose(f_phonemes_out);
+    
+    return 0;
+  }
+
   long set_voice(
         const char* aName,
         const char* aLang=NULL,

emscripten/espeakng_glue.idl

@@ -52,6 +52,7 @@ interface espeak_EVENT {
 interface eSpeakNGWorker {
   void eSpeakNGWorker();
   void synth_(DOMString aText, VoidPtr aCallback);
+  long synth_ipa_(DOMString aText, DOMString virtualFileName);
   long getSizeOfEventStruct_();
   long set_voice(DOMString aName, DOMString aLang, optional octet gender=0, optional octet age=0, optional octet aVariant=0);
   [Const] attribute espeak_VOICE[] voices;

emscripten/js/demo.js

@@ -179,10 +179,12 @@ function speak() {
   pusher.connect(ctx.destination);
   console.log('  Creating pusher... done');
 
+  var user_text = document.getElementById('texttospeak').value;
+  
   // actual synthesis
   console.log('  Calling synthesize...');
   tts.synthesize(
-    document.getElementById('texttospeak').value,
+    user_text,
     function cb(samples, events) {
       //console.log('  Inside synt cb');
       if (!samples) {
@@ -203,11 +205,32 @@ function speak() {
       //console.log('  Leaving synt cb');
     } // end of function cb
   ); // end of tts.synthesize()
-  console.log('  Calling synthesize... done');
-
+  console.log('  Calling synthesize... done');  
   console.log('Leaving speak()');
+
 } // end of speak()
 
+function ipa() {
+  
+  console.log("Synthesizing ipa ... ");
+  var ts = new Date();
+  var user_text = document.getElementById('texttospeak').value;
+
+ //user_text = user_text.repeat(50);
+  
+  tts.set_voice(document.getElementById('voice').value);
+  tts.synthesize_ipa(user_text, function(result) { 
+    var te = new Date();
+    document.getElementById('ipaarea').value = result.ipa;
+    console.log("Ipa synthesis done in " + (te-ts) + " ms.")
+  });
+}
+
+function speakAndIpa() {
+  speak();
+  ipa();
+}
+
 function resetPitch() {
   document.getElementById('pitch').value = 50;
 }
@@ -254,5 +277,6 @@ function initializeDemo() {
       console.log('Leaving cb1');
     } // end of function cb1
   );
+    
   console.log('Creating eSpeakNG instance... done');
 }

emscripten/js/espeakng.js

@@ -63,7 +63,8 @@ for (var method of [
     'set_rate',
     'set_pitch',
     'set_voice',
-    'synthesize'
+    'synthesize',
+    'synthesize_ipa'
     ]) {
   eSpeakNG.prototype[method] = _createAsyncMethod(method);
 }

emscripten/post.js

@@ -93,6 +93,28 @@ eSpeakNGWorker.prototype.synthesize = function (aText, aCallback) {
   Runtime.removeFunction(fp);
 };
 
+eSpeakNGWorker.prototype.synthesize_ipa = function (aText, aCallback) {
+  
+  // Use a unique temp file for the worker. Avoid collisions, just in case.
+  var ipaVirtualFileName = "espeak-ng-ipa-tmp-"  + Math.random().toString().substring(2);
+  
+  var res = "";
+  var code = this.synth_ipa_(aText, ipaVirtualFileName);
+
+  if(code == 0)
+    res = FS.readFile(ipaVirtualFileName, { encoding: 'utf8' })
+    
+  // Clean up the tmp file
+  FS.unlink(ipaVirtualFileName);
+    
+  var ret = {
+    code: code,
+    ipa: res
+  }
+  
+  return ret;
+};
+
 // Make this a worker
 
 if (typeof WorkerGlobalScope !== 'undefined') {
@@ -104,6 +126,7 @@ if (typeof WorkerGlobalScope !== 'undefined') {
   });
 
   onmessage = function(e) {
+    
     if (!worker) {
       throw 'eSpeakNGWorker worker not initialized';
     }

espeak-ng-data/lang/inc/sd

@@ -1,2 +1,3 @@
 name Sindhi
 language sd
+maintainer Ejaz Shah <[email protected]>

espeak-ng-data/lang/poz/mi

@@ -1,3 +1,4 @@
+name Māori
 language mi
 status testing
 

m4/ax_check_compile_flag.m4

@@ -0,0 +1,75 @@
+# ===========================================================================
+#  https://www.gnu.org/software/autoconf-archive/ax_check_compile_flag.html
+# ===========================================================================
+#
+# SYNOPSIS
+#
+#   AX_CHECK_COMPILE_FLAG(FLAG, [ACTION-SUCCESS], [ACTION-FAILURE], [EXTRA-FLAGS], [INPUT])
+#
+# DESCRIPTION
+#
+#   Check whether the given FLAG works with the current language's compiler
+#   or gives an error.  (Warnings, however, are ignored)
+#
+#   ACTION-SUCCESS/ACTION-FAILURE are shell commands to execute on
+#   success/failure.
+#
+#   If EXTRA-FLAGS is defined, it is added to the current language's default
+#   flags (e.g. CFLAGS) when the check is done.  The check is thus made with
+#   the flags: "CFLAGS EXTRA-FLAGS FLAG".  This can for example be used to
+#   force the compiler to issue an error when a bad flag is given.
+#
+#   INPUT gives an alternative input source to AC_COMPILE_IFELSE.
+#
+#   NOTE: Implementation based on AX_CFLAGS_GCC_OPTION. Please keep this
+#   macro in sync with AX_CHECK_{PREPROC,LINK}_FLAG.
+#
+# LICENSE
+#
+#   Copyright (c) 2008 Guido U. Draheim <[email protected]>
+#   Copyright (c) 2011 Maarten Bosmans <[email protected]>
+#
+#   This program is free software: you can redistribute it and/or modify it
+#   under the terms of the GNU General Public License as published by the
+#   Free Software Foundation, either version 3 of the License, or (at your
+#   option) any later version.
+#
+#   This program is distributed in the hope that it will be useful, but
+#   WITHOUT ANY WARRANTY; without even the implied warranty of
+#   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
+#   Public License for more details.
+#
+#   You should have received a copy of the GNU General Public License along
+#   with this program. If not, see <https://www.gnu.org/licenses/>.
+#
+#   As a special exception, the respective Autoconf Macro's copyright owner
+#   gives unlimited permission to copy, distribute and modify the configure
+#   scripts that are the output of Autoconf when processing the Macro. You
+#   need not follow the terms of the GNU General Public License when using
+#   or distributing such scripts, even though portions of the text of the
+#   Macro appear in them. The GNU General Public License (GPL) does govern
+#   all other use of the material that constitutes the Autoconf Macro.
+#
+#   This special exception to the GPL applies to versions of the Autoconf
+#   Macro released by the Autoconf Archive. When you make and distribute a
+#   modified version of the Autoconf Macro, you may extend this special
+#   exception to the GPL to apply to your modified version as well.
+
+#serial 5
+
+AC_DEFUN([AX_CHECK_COMPILE_FLAG],
+[AC_PREREQ(2.64)dnl for _AC_LANG_PREFIX and AS_VAR_IF
+AS_VAR_PUSHDEF([CACHEVAR],[ax_cv_check_[]_AC_LANG_ABBREV[]flags_$4_$1])dnl
+AC_CACHE_CHECK([whether _AC_LANG compiler accepts $1], CACHEVAR, [
+  ax_check_save_flags=$[]_AC_LANG_PREFIX[]FLAGS
+  _AC_LANG_PREFIX[]FLAGS="$[]_AC_LANG_PREFIX[]FLAGS $4 $1"
+  AC_COMPILE_IFELSE([m4_default([$5],[AC_LANG_PROGRAM()])],
+    [AS_VAR_SET(CACHEVAR,[yes])],
+    [AS_VAR_SET(CACHEVAR,[no])])
+  _AC_LANG_PREFIX[]FLAGS=$ax_check_save_flags])
+AS_VAR_IF(CACHEVAR,yes,
+  [m4_default([$2], :)],
+  [m4_default([$3], :)])
+AS_VAR_POPDEF([CACHEVAR])dnl
+])dnl AX_CHECK_COMPILE_FLAGS
+

phsource/ph_latvian

@@ -198,6 +198,25 @@ phoneme n- // more voiced variant of n
   FMT(n/n-syl)
 endphoneme
 
+phoneme h // more quiet version of standard h
+  vls glt apr
+  lengthmod 3
+  ipa h
+
+  IF nextPh(#a) THEN
+    WAV(h/ha, 20)
+  ELIF nextPh(#e) THEN
+    WAV(h/he, 20)
+  ELIF nextPh(#i) THEN
+    WAV(h/hi, 20)
+  ELIF nextPh(#o) THEN
+    WAV(h/ho, 20)
+  ELIF nextPh(#u) THEN
+    WAV(h/hu, 20)
+  ENDIF
+  WAV(h/h_, 20)  // no vowel following
+endphoneme
+
 phoneme R // Latvian hard r
   liquid trl
   lengthmod 6
@@ -336,15 +355,6 @@ phoneme } // Tiny, 5ms non-linked pause (used between phonemes in word)
   length 5
 endphoneme
 
-// TODO should use this with correct lengthening
-//phoneme l
-//  liquid
-//  unstressed
-//  dnt
-//  length 150
-//  FMT(vwl_hi/l-voc)
-//endphoneme
-
 phoneme s
   vls alv frc sib
   voicingswitch z
@@ -422,4 +432,3 @@ phoneme l
 endphoneme
 
 
-

src/espeak-ng.c

@@ -126,7 +126,7 @@ FILE *f_wavfile = NULL;
 char filetype[5];
 char wavefile[200];
 
-void DisplayVoices(FILE *f_out, char *language)
+static void DisplayVoices(FILE *f_out, char *language)
 {
 	int ix;
 	const char *p;
@@ -197,7 +197,7 @@ static void Write4Bytes(FILE *f, int value)
 	}
 }
 
-int OpenWavFile(char *path, int rate)
+static int OpenWavFile(char *path, int rate)
 {
 	static unsigned char wave_hdr[44] = {
 		'R', 'I', 'F', 'F', 0x24, 0xf0, 0xff, 0x7f, 'W', 'A', 'V', 'E', 'f', 'm', 't', ' ',
@@ -598,7 +598,7 @@ int main(int argc, char **argv)
 	}
 
 	if (voicename[0] == 0)
-		strcpy(voicename, "en");
+		strcpy(voicename, ESPEAKNG_DEFAULT_VOICE);
 
 	result = espeak_ng_SetVoiceByName(voicename);
 	if (result != ENS_OK) {

src/include/compat/wchar.h

@@ -32,7 +32,4 @@
 
 #include <ucd/ucd.h>
 
-#define towlower ucd_tolower
-#define towupper ucd_toupper
-
 #endif

src/include/espeak-ng/encoding.h

@@ -17,6 +17,8 @@
 #ifndef ESPEAK_NG_ENCODING_H
 #define ESPEAK_NG_ENCODING_H
 
+#include <stdint.h>
+
 #ifdef __cplusplus
 extern "C"
 {

src/include/espeak-ng/espeak_ng.h

@@ -36,6 +36,8 @@ extern "C"
 #define ESPEAK_NG_API
 #endif
 
+#define ESPEAKNG_DEFAULT_VOICE "en"
+
 typedef enum {
 	ENS_GROUP_MASK               = 0x70000000,
 	ENS_GROUP_ERRNO              = 0x00000000, /* Values 0-255 map to errno error codes. */

src/include/espeak-ng/speak_lib.h

@@ -38,7 +38,7 @@
 #define ESPEAK_API
 #endif
 
-#define ESPEAK_API_REVISION  10
+#define ESPEAK_API_REVISION  12
 /*
 Revision 2
    Added parameter "options" to eSpeakInitialize()
@@ -68,6 +68,12 @@ Revision 9  30.May.2013
 Revision 10 29.Aug.2014
   Changed phonememode parameter to espeak_TextToPhonemes() and espeak_SetPhonemeTrace
 
+Revision 11 (espeak-ng)
+  Made ESPEAK_API import/export symbols correctly on Windows.
+
+Revision 12 (espeak-ng)
+  Exposed espeak_SetPhonemeCallback. This is available in eSpeak, but was not exposed in this header.
+
 */
          /********************/
          /*  Initialization  */
@@ -260,6 +266,11 @@ int UriCallback(int type, const char *uri, const char *base);
              occurs.  The calling program can then play the sound at that point.
 */
 
+#ifdef __cplusplus
+extern "C"
+#endif
+ESPEAK_API void espeak_SetPhonemeCallback(int (*PhonemeCallback)(const char *));
+
 
          /********************/
          /*    Synthesis     */

src/include/espeak-ng/tokenizer.h

@@ -1,75 +0,0 @@
-/*
- * Copyright (C) 2017 Reece H. Dunn
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, see: <http://www.gnu.org/licenses/>.
- */
-#ifndef ESPEAK_NG_TOKENIZER_H
-#define ESPEAK_NG_TOKENIZER_H
-
-#ifdef __cplusplus
-extern "C"
-{
-#endif
-
-typedef struct espeak_ng_TOKENIZER_ espeak_ng_TOKENIZER;
-
-ESPEAK_NG_API espeak_ng_TOKENIZER *
-create_tokenizer(void);
-
-ESPEAK_NG_API void
-destroy_tokenizer(espeak_ng_TOKENIZER *tokenizer);
-
-typedef enum
-{
-	ESPEAKNG_TOKENIZER_OPTION_TEXT = 0,
-} espeak_ng_TOKENIZER_OPTIONS;
-
-ESPEAK_NG_API int
-tokenizer_reset(espeak_ng_TOKENIZER *tokenizer,
-                espeak_ng_TEXT_DECODER *decoder,
-                espeak_ng_TOKENIZER_OPTIONS options);
-
-typedef enum
-{
-	ESPEAKNG_TOKEN_END_OF_BUFFER,
-	ESPEAKNG_TOKEN_UNKNOWN,
-	ESPEAKNG_TOKEN_NEWLINE,
-	ESPEAKNG_TOKEN_PARAGRAPH,
-	ESPEAKNG_TOKEN_WHITESPACE,
-	ESPEAKNG_TOKEN_WORD_UPPERCASE,
-	ESPEAKNG_TOKEN_WORD_LOWERCASE,
-	ESPEAKNG_TOKEN_WORD_MIXEDCASE,
-	ESPEAKNG_TOKEN_WORD_CAPITALIZED,
-	ESPEAKNG_TOKEN_FULL_STOP,
-	ESPEAKNG_TOKEN_QUESTION_MARK,
-	ESPEAKNG_TOKEN_EXCLAMATION_MARK,
-	ESPEAKNG_TOKEN_COMMA,
-	ESPEAKNG_TOKEN_COLON,
-	ESPEAKNG_TOKEN_SEMICOLON,
-	ESPEAKNG_TOKEN_ELLIPSIS,
-	ESPEAKNG_TOKEN_PUNCTUATION,
-	ESPEAKNG_TOKEN_SYMBOL,
-} espeak_ng_TOKEN_TYPE;
-
-ESPEAK_NG_API espeak_ng_TOKEN_TYPE
-tokenizer_read_next_token(espeak_ng_TOKENIZER *tokenizer);
-
-ESPEAK_NG_API const char *
-tokenizer_get_token_text(espeak_ng_TOKENIZER *tokenizer);
-
-#ifdef __cplusplus
-}
-#endif
-
-#endif

src/libespeak-ng/compiledata.c

@@ -33,13 +33,15 @@
 
 #include <espeak-ng/espeak_ng.h>
 #include <espeak-ng/speak_lib.h>
+#include <espeak-ng/encoding.h>
 
 #include "error.h"
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
 #include "spect.h"
+#include "translate.h"
 
 typedef struct {
 	unsigned int value;
@@ -50,9 +52,6 @@ NAMETAB *manifest = NULL;
 int n_manifest;
 char phsrc[sizeof(path_home)+40]; // Source: path to the 'phonemes' source file.
 
-extern ESPEAK_NG_API int utf8_in(int *c, const char *buf);
-extern int utf8_out(unsigned int c, char *buf);
-
 typedef struct {
 	const char *mnem;
 	int type;
@@ -722,7 +721,7 @@ static void unget_char(unsigned int c)
 		linenum--;
 }
 
-int CheckNextChar()
+static int CheckNextChar()
 {
 	int c;
 	while (((c = get_char()) == ' ') || (c == '\t'))
@@ -893,7 +892,7 @@ static int Range(int value, int divide, int min, int max)
 	return value - min;
 }
 
-int CompileVowelTransition(int which)
+static int CompileVowelTransition(int which)
 {
 	// Compile a vowel transition
 	int key;
@@ -999,7 +998,7 @@ int CompileVowelTransition(int which)
 	return 0;
 }
 
-espeak_ng_STATUS LoadSpect(const char *path, int control, int *addr)
+static espeak_ng_STATUS LoadSpect(const char *path, int control, int *addr)
 {
 	SpectSeq *spectseq;
 	int peak;
@@ -1401,15 +1400,17 @@ static int LoadEnvelope2(FILE *f, const char *fname)
 			n_points++;
 		}
 	}
-	env_x[n_points] = env_x[n_points-1];
-	env_y[n_points] = env_y[n_points-1];
+	if (n_points > 0) {
+		env_x[n_points] = env_x[n_points-1];
+		env_y[n_points] = env_y[n_points-1];
+	}
 
 	ix = -1;
 	ix2 = 0;
-	for (x = 0; x < ENV_LEN; x++) {
-		if (x > env_x[ix+4])
+	if (n_points > 0) for (x = 0; x < ENV_LEN; x++) {
+		if (n_points > 3 && x > env_x[ix+4])
 			ix++;
-		if (x >= env_x[ix2+1])
+		if (n_points > 2 && x >= env_x[ix2+1])
 			ix2++;
 
 		if (env_lin[ix2] > 0) {
@@ -1612,7 +1613,7 @@ static void CompileSound(int keyword, int isvowel)
    =8         data = stress bitmap
    =9         special tests
  */
-int CompileIf(int elif)
+static int CompileIf(int elif)
 {
 	int key;
 	int finish = 0;
@@ -1721,7 +1722,7 @@ int CompileIf(int elif)
 	return 0;
 }
 
-void FillThen(int add)
+static void FillThen(int add)
 {
 	USHORT *p;
 	int offset;
@@ -1749,7 +1750,7 @@ void FillThen(int add)
 	then_count = 0;
 }
 
-int CompileElse(void)
+static int CompileElse(void)
 {
 	USHORT *ref;
 	USHORT *p;
@@ -1776,7 +1777,7 @@ int CompileElse(void)
 	return 0;
 }
 
-int CompileElif(void)
+static int CompileElif(void)
 {
 	if (if_level < 1) {
 		error("ELIF not expected");
@@ -1788,7 +1789,7 @@ int CompileElif(void)
 	return 0;
 }
 
-int CompileEndif(void)
+static int CompileEndif(void)
 {
 	USHORT *p;
 	int chain;
@@ -1951,7 +1952,7 @@ static void DecThenCount()
 		then_count--;
 }
 
-int CompilePhoneme(int compile_phoneme)
+static int CompilePhoneme(int compile_phoneme)
 {
 	int endphoneme = 0;
 	int keyword;
@@ -2231,11 +2232,11 @@ int CompilePhoneme(int compile_phoneme)
 				CompileToneSpec();
 				break;
 			case kCONTINUE:
-				*prog_out++ = OPCODE_CONTINUE;
+				*prog_out++ = INSTN_CONTINUE;
 				DecThenCount();
 				break;
 			case kRETURN:
-				*prog_out++ = OPCODE_RETURN;
+				*prog_out++ = INSTN_RETURN;
 				DecThenCount();
 				break;
 			case kINCLUDE:
@@ -2248,7 +2249,7 @@ int CompilePhoneme(int compile_phoneme)
 				if (if_level > 0)
 					error("Missing ENDIF");
 				if ((prog_out > prog_buf) && (if_stack[0].returned == 0))
-					*prog_out++ = OPCODE_RETURN;
+					*prog_out++ = INSTN_RETURN;
 				break;
 			}
 			break;
@@ -2357,8 +2358,6 @@ static void WritePhonemeTables()
 static void EndPhonemeTable()
 {
 	int ix;
-	int *pw;
-	int length;
 
 	if (n_phoneme_tabs == 0)
 		return;
@@ -2697,7 +2696,7 @@ MNEM_TAB envelope_names[] = {
 	{ NULL, -1 }
 };
 
-int LookupEnvelopeName(const char *name)
+static int LookupEnvelopeName(const char *name)
 {
 	return LookupMnem(envelope_names, name);
 }
@@ -2717,7 +2716,7 @@ espeak_ng_STATUS espeak_ng_CompileIntonation(FILE *log, espeak_ng_ERROR_CONTEXT
 	int done_onset = 0;
 	int done_last = 0;
 	int n_preset_tunes = 0;
-	int found;
+	int found = 0;
 	int tune_number = 0;
 	FILE *f_out;
 	TUNE *tune_data;
@@ -2825,6 +2824,7 @@ espeak_ng_STATUS espeak_ng_CompileIntonation(FILE *log, espeak_ng_ERROR_CONTEXT
 				error("Bad tune name: '%s;", new_tune.name);
 			break;
 		case kENDTUNE:
+			if (!found) continue;
 			if (done_onset == 0) {
 				new_tune.unstr_start[0] = new_tune.unstr_start[1];
 				new_tune.unstr_end[0] = new_tune.unstr_end[1];

src/libespeak-ng/compiledict.c

@@ -35,11 +35,11 @@
 #include "error.h"
 #include "speech.h"
 #include "phoneme.h"
+#include "voice.h"
 #include "synthesize.h"
 #include "translate.h"
 
 extern void Write4Bytes(FILE *f, int value);
-int HashDictionary(const char *string);
 
 static FILE *f_log = NULL;
 extern char *dir_dictionary;
@@ -52,8 +52,15 @@ static int text_mode = 0;
 static int debug_flag = 0;
 static int error_need_dictionary = 0;
 
-static int hash_counts[N_HASH_DICT];
+// A hash chain is a linked-list of hash chain entry objects:
+//     struct hash_chain_entry {
+//         hash_chain_entry *next_entry;
+//         // dict_line output from compile_line:
+//         uint8_t length;
+//         char contents[length];
+//     };
 static char *hash_chains[N_HASH_DICT];
+
 static char letterGroupsDefined[N_LETTER_GROUPS];
 
 MNEM_TAB mnem_rules[] = {
@@ -164,16 +171,6 @@ typedef struct {
 	int group3_ix;
 } RGROUP;
 
-int isspace2(unsigned int c)
-{
-	// can't use isspace() because on Windows, isspace(0xe1) gives TRUE !
-	int c2;
-
-	if (((c2 = (c & 0xff)) == 0) || (c > ' '))
-		return 0;
-	return 1;
-}
-
 void print_dictionary_flags(unsigned int *flags, char *buf, int buf_len)
 {
 	int stress;
@@ -411,8 +408,8 @@ static int compile_line(char *linebuf, char *dict_line, int n_dict_line, int *ha
 
 	step = LINE_PARSER_WORD;
 
-	c = 0;
-	while (c != '\n') {
+	c = *p;
+	while (c != '\n' && c != '\0') {
 		c = *p;
 
 		if ((c == '?') && (step == 0)) {
@@ -649,7 +646,7 @@ static int compile_line(char *linebuf, char *dict_line, int n_dict_line, int *ha
 			length += ix;
 		}
 	}
-	dict_line[0] = length;
+	*((uint8_t *)dict_line) = (uint8_t)length;
 
 	return length;
 }
@@ -669,7 +666,6 @@ static void compile_dictlist_start(void)
 			p = p2;
 		}
 		hash_chains[ix] = NULL;
-		hash_counts[ix] = 0;
 	}
 }
 
@@ -682,10 +678,9 @@ static void compile_dictlist_end(FILE *f_out)
 
 	for (hash = 0; hash < N_HASH_DICT; hash++) {
 		p = hash_chains[hash];
-		hash_counts[hash] = (int)ftell(f_out);
 
 		while (p != NULL) {
-			length = *(p+sizeof(char *));
+			length = *(uint8_t *)(p+sizeof(char *));
 			fwrite(p+sizeof(char *), length, 1, f_out);
 			memcpy(&p, p, sizeof(char *));
 		}
@@ -702,7 +697,7 @@ static int compile_dictlist_file(const char *path, const char *filename)
 	FILE *f_in;
 	char buf[200];
 	char fname[sizeof(path_home)+45];
-	char dict_line[200];
+	char dict_line[256]; // length is uint8_t, so an entry can't take up more than 256 bytes
 
 	text_mode = 0;
 
@@ -725,8 +720,6 @@ static int compile_dictlist_file(const char *path, const char *filename)
 		length = compile_line(buf, dict_line, sizeof(dict_line), &hash);
 		if (length == 0)  continue; // blank line
 
-		hash_counts[hash]++;
-
 		p = (char *)malloc(length+sizeof(char *));
 		if (p == NULL) {
 			if (f_log != NULL) {
@@ -738,6 +731,7 @@ static int compile_dictlist_file(const char *path, const char *filename)
 
 		memcpy(p, &hash_chains[hash], sizeof(char *));
 		hash_chains[hash] = p;
+		// NOTE: dict_line[0] is the entry length (0-255)
 		memcpy(p+sizeof(char *), dict_line, length);
 		count++;
 	}
@@ -758,7 +752,7 @@ static int group3_ix;
 
 #define N_RULES 3000 // max rules for each group
 
-int isHexDigit(int c)
+static int isHexDigit(int c)
 {
 	if ((c >= '0') && (c <= '9'))
 		return c - '0';
@@ -1165,7 +1159,7 @@ static char *compile_rule(char *input)
 	return prule;
 }
 
-int __cdecl string_sorter(char **a, char **b)
+static int __cdecl string_sorter(char **a, char **b)
 {
 	char *pa, *pb;
 	int ix;

src/libespeak-ng/compilembrola.c

@@ -31,6 +31,7 @@
 #include "error.h"
 #include "phoneme.h"
 #include "speech.h"
+#include "voice.h"
 #include "synthesize.h"
 
 static const char *basename(const char *filename)

src/libespeak-ng/dictionary.c

@@ -33,15 +33,13 @@
 
 #include "speech.h"
 #include "phoneme.h"
+#include "voice.h"
 #include "synthesize.h"
 #include "translate.h"
 
 int dictionary_skipwords;
 char dictionary_name[40];
 
-extern void print_dictionary_flags(unsigned int *flags, char *buf, int buf_len);
-extern char *DecodeRule(const char *group_chars, int group_length, char *rule, int control);
-
 // accented characters which indicate (in some languages) the start of a separate syllable
 static const unsigned short diereses_list[7] = { 0xe4, 0xeb, 0xef, 0xf6, 0xfc, 0xff, 0 };
 
@@ -198,7 +196,7 @@ int LoadDictionary(Translator *tr, const char *name, int no_error)
 	int *pw;
 	int length;
 	FILE *f;
-	unsigned int size;
+	int size;
 	char fname[sizeof(path_home)+20];
 
 	strncpy(dictionary_name, name, 40); // currently loaded dictionary name
@@ -254,7 +252,7 @@ int LoadDictionary(Translator *tr, const char *name, int no_error)
 
 	for (hash = 0; hash < N_HASH_DICT; hash++) {
 		tr->dict_hashtab[hash] = p;
-		while ((length = *p) != 0)
+		while ((length = *(uint8_t *)p) != 0)
 			p += length;
 		p++; // skip over the zero which terminates the list for this hash value
 	}
@@ -1571,7 +1569,6 @@ static void MatchRule(Translator *tr, char *word[], char *word_start, int group_
 	int letter_w;         // current letter, wide character
 	int last_letter_w;    // last letter, wide character
 	int letter_xbytes;    // number of extra bytes of multibyte character (num bytes - 1)
-	unsigned char last_letter;
 
 	char *pre_ptr;
 	char *post_ptr;       // pointer to first character after group
@@ -1627,7 +1624,6 @@ static void MatchRule(Translator *tr, char *word[], char *word_start, int group_
 		unpron_ignore = word_flags & FLAG_UNPRON_TEST;
 		match_type = 0;
 		consumed = 0;
-		letter = 0;
 		letter_w = 0;
 		distance_right = -6; // used to reduce points for matches further away the current letter
 		distance_left = -2;
@@ -1732,7 +1728,6 @@ static void MatchRule(Translator *tr, char *word[], char *word_start, int group_
 				distance_right += 6;
 				if (distance_right > 18)
 					distance_right = 19;
-				last_letter = letter;
 				last_letter_w = letter_w;
 				letter_xbytes = utf8_in(&letter_w, post_ptr)-1;
 				letter = *post_ptr++;
@@ -1939,7 +1934,6 @@ static void MatchRule(Translator *tr, char *word[], char *word_start, int group_
 				if (distance_left > 18)
 					distance_left = 19;
 
-				last_letter = *pre_ptr;
 				utf8_in(&last_letter_w, pre_ptr);
 				pre_ptr--;
 				letter_xbytes = utf8_in2(&letter_w, pre_ptr, 1)-1;
@@ -2719,7 +2713,7 @@ static const char *LookupDict2(Translator *tr, const char *word, const char *wor
 			continue;
 		}
 
-		if ((dictionary_flags2 & FLAG_ATSTART) && !(wtab->flags & FLAG_FIRST_WORD)) {
+		if ((dictionary_flags2 & FLAG_ATSTART) && !(wflags & FLAG_FIRST_WORD)) {
 			// only use this pronunciation if it's the first word of a clause
 			continue;
 		}
@@ -2894,7 +2888,7 @@ int LookupDictList(Translator *tr, char **wordptr, char *ph_out, unsigned int *f
 		found = word2 + len;
 	}
 
-	if (found == 0) {
+	if (found == 0 && length >= 2) {
 		ph_out[0] = 0;
 
 		// try modifications to find a recognised word
@@ -3003,7 +2997,7 @@ int RemoveEnding(Translator *tr, char *word, int end_type, char *word_copy)
 	int end_flags;
 	const char *p;
 	int len;
-	char ending[50];
+	char ending[50] = {0};
 
 	// these lists are language specific, but are only relevent if the 'e' suffix flag is used
 	static const char *add_e_exceptions[] = {

src/libespeak-ng/encoding.c

@@ -26,43 +26,10 @@
 #include <espeak-ng/encoding.h>
 
 #include "speech.h"
-
-#define LEADING_2_BITS 0xC0 // 0b11000000
-#define UTF8_TAIL_BITS 0x80 // 0b10000000
-
-int utf8_out(unsigned int c, char *buf)
-{
-	// write a unicode character into a buffer as utf8
-	// returns the number of bytes written
-
-	int n_bytes;
-	int j;
-	int shift;
-	static char unsigned code[4] = { 0, 0xc0, 0xe0, 0xf0 };
-
-	if (c < 0x80) {
-		buf[0] = c;
-		return 1;
-	}
-	if (c >= 0x110000) {
-		buf[0] = ' '; // out of range character code
-		return 1;
-	}
-	if (c < 0x0800)
-		n_bytes = 1;
-	else if (c < 0x10000)
-		n_bytes = 2;
-	else
-		n_bytes = 3;
-
-	shift = 6*n_bytes;
-	buf[0] = code[n_bytes] | (c >> shift);
-	for (j = 0; j < n_bytes; j++) {
-		shift -= 6;
-		buf[j+1] = 0x80 + ((c >> shift) & 0x3f);
-	}
-	return n_bytes+1;
-}
+#include "phoneme.h"
+#include "voice.h"
+#include "synthesize.h"
+#include "translate.h"
 
 // http://www.iana.org/assignments/character-sets/character-sets.xhtml
 MNEM_TAB mnem_encoding[] = {
@@ -654,6 +621,7 @@ string_decoder_getc_auto(espeak_ng_TEXT_DECODER *decoder)
 static uint32_t
 null_decoder_getc(espeak_ng_TEXT_DECODER *decoder)
 {
+	(void)decoder; // unused parameter
 	return 0;
 }
 

src/libespeak-ng/error.c

@@ -29,6 +29,7 @@
 #include "error.h"
 #include "speech.h"
 #include "phoneme.h"
+#include "voice.h"
 #include "synthesize.h"
 
 espeak_ng_STATUS

src/libespeak-ng/espeak_api.c

@@ -29,6 +29,7 @@
 
 #include "speech.h"
 #include "phoneme.h"
+#include "voice.h"
 #include "synthesize.h"
 #include "translate.h"
 #include "event.h"

src/libespeak-ng/event.c

@@ -66,8 +66,8 @@ static node *head = NULL;
 static node *tail = NULL;
 static int node_counter = 0;
 static espeak_ng_STATUS push(void *data);
-static void *pop();
-static void init();
+static void *pop(void);
+static void init(void);
 static void *polling_thread(void *);
 
 void event_set_callback(t_espeak_callback *SynthCallback)
@@ -257,7 +257,7 @@ static void *polling_thread(void *p)
 	while (1) {
 		int a_stop_is_required = 0;
 
-		int a_status = pthread_mutex_lock(&my_mutex);
+		(void)pthread_mutex_lock(&my_mutex);
 		my_event_is_running = 0;
 
 		while (my_start_is_required == 0) {
@@ -284,16 +284,16 @@ static void *polling_thread(void *p)
 				event->user_data = NULL;
 			}
 
-			a_status = pthread_mutex_lock(&my_mutex);
+			(void)pthread_mutex_lock(&my_mutex);
 			event_delete((espeak_EVENT *)pop());
 			a_stop_is_required = my_stop_is_required;
 			if (a_stop_is_required > 0)
 				my_stop_is_required = 0;
 
-			a_status = pthread_mutex_unlock(&my_mutex);
+			(void)pthread_mutex_unlock(&my_mutex);
 		}
 
-		a_status = pthread_mutex_lock(&my_mutex);
+		(void)pthread_mutex_lock(&my_mutex);
 
 		my_event_is_running = 0;
 
@@ -303,7 +303,7 @@ static void *polling_thread(void *p)
 				my_stop_is_required = 0;
 		}
 
-		a_status = pthread_mutex_unlock(&my_mutex);
+		(void)pthread_mutex_unlock(&my_mutex);
 
 		if (a_stop_is_required > 0) {
 			// no mutex required since the stop command is synchronous
@@ -311,10 +311,10 @@ static void *polling_thread(void *p)
 			init();
 
 			// acknowledge the stop request
-			espeak_ng_STATUS a_status = pthread_mutex_lock(&my_mutex);
+			(void)pthread_mutex_lock(&my_mutex);
 			my_stop_is_acknowledged = 1;
-			a_status = pthread_cond_signal(&my_cond_stop_is_acknowledged);
-			a_status = pthread_mutex_unlock(&my_mutex);
+			(void)pthread_cond_signal(&my_cond_stop_is_acknowledged);
+			(void)pthread_mutex_unlock(&my_mutex);
 		}
 	}
 

src/libespeak-ng/event.h

@@ -52,14 +52,14 @@ void event_init(void);
 void event_set_callback(t_espeak_callback *cb);
 
 // Clear any pending event.
-espeak_ng_STATUS event_clear_all();
+espeak_ng_STATUS event_clear_all(void);
 
 // Declare a future event
 espeak_ng_STATUS event_declare(espeak_EVENT *event);
 
 // Terminate the event component.
 // Last function to be called.
-void event_terminate();
+void event_terminate(void);
 
 // general functions
 void clock_gettime2(struct timespec *ts);

src/libespeak-ng/fifo.c

@@ -58,7 +58,7 @@ static int my_stop_is_acknowledged = 0;
 static void *say_thread(void *);
 
 static espeak_ng_STATUS push(t_espeak_command *the_command);
-static t_espeak_command *pop();
+static t_espeak_command *pop(void);
 static void init(int process_parameters);
 static int node_counter = 0;
 

src/libespeak-ng/fifo.h

@@ -29,7 +29,7 @@ extern "C"
 
 // Initialize the fifo component.
 // First function to be called.
-void fifo_init();
+void fifo_init(void);
 
 // Add an espeak command.
 //
@@ -44,15 +44,15 @@ espeak_ng_STATUS fifo_add_command(t_espeak_command *c);
 espeak_ng_STATUS fifo_add_commands(t_espeak_command *c1, t_espeak_command *c2);
 
 // The current running command must be stopped and the awaiting commands are cleared.
-espeak_ng_STATUS fifo_stop();
+espeak_ng_STATUS fifo_stop(void);
 
 // Is there a running command?
 // Returns 1 if yes; 0 otherwise.
-int fifo_is_busy();
+int fifo_is_busy(void);
 
 // Terminate the fifo component.
 // Last function to be called.
-void fifo_terminate();
+void fifo_terminate(void);
 
 // Indicates if the running command is still enabled.
 //
@@ -61,7 +61,7 @@ void fifo_terminate();
 // stopping speech as soon as a cancel command is applied.
 //
 // Returns 1 if yes, or 0 otherwise.
-int fifo_is_command_enabled();
+int fifo_is_command_enabled(void);
 
 #ifdef __cplusplus
 }

src/libespeak-ng/ieee80.c

@@ -49,15 +49,10 @@
 
 #include	<stdio.h>
 #include	<math.h>
+#include	"ieee80.h"
 
 typedef float Single;
 
-#ifndef applec
- typedef double defdouble;
-#else applec
- typedef long double defdouble;
-#endif applec
-
 #ifndef THINK_C
  typedef double Double;
 #else THINK_C

src/libespeak-ng/ieee80.h

@@ -0,0 +1,62 @@
+/* Copyright (C) 1989-1991 Apple Computer, Inc.
+ *
+ * All rights reserved.
+ *
+ * Warranty Information
+ *  Even though Apple has reviewed this software, Apple makes no warranty
+ *  or representation, either express or implied, with respect to this
+ *  software, its quality, accuracy, merchantability, or fitness for a
+ *  particular purpose.  As a result, this software is provided "as is,"
+ *  and you, its user, are assuming the entire risk as to its quality
+ *  and accuracy.
+ *
+ * This code may be used and freely distributed as long as it includes
+ * this copyright notice and the above warranty information.
+ *
+ * Machine-independent I/O routines for IEEE floating-point numbers.
+ *
+ * NaN's and infinities are converted to HUGE_VAL or HUGE, which
+ * happens to be infinity on IEEE machines.  Unfortunately, it is
+ * impossible to preserve NaN's in a machine-independent way.
+ * Infinities are, however, preserved on IEEE machines.
+ *
+ * These routines have been tested on the following machines:
+ *	Apple Macintosh, MPW 3.1 C compiler
+ *	Apple Macintosh, THINK C compiler
+ *	Silicon Graphics IRIS, MIPS compiler
+ *	Cray X/MP and Y/MP
+ *	Digital Equipment VAX
+ *	Sequent Balance (Multiprocesor 386)
+ *	NeXT
+ *
+ *
+ * Implemented by Malcolm Slaney and Ken Turkowski.
+ *
+ * Malcolm Slaney contributions during 1988-1990 include big- and little-
+ * endian file I/O, conversion to and from Motorola's extended 80-bit
+ * floating-point format, and conversions to and from IEEE single-
+ * precision floating-point format.
+ *
+ * In 1991, Ken Turkowski implemented the conversions to and from
+ * IEEE double-precision format, added more precision to the extended
+ * conversions, and accommodated conversions involving +/- infinity,
+ * NaN's, and denormalized numbers.
+ */
+#ifndef IEEE_H
+#define IEEE_H
+
+#ifndef applec
+ typedef double defdouble;
+#else applec
+ typedef long double defdouble;
+#endif applec
+
+defdouble ConvertFromIeeeSingle(  char *bytes);
+defdouble ConvertFromIeeeDouble(  char *bytes);
+defdouble ConvertFromIeeeExtended(char *bytes);
+
+void ConvertToIeeeSingle(  defdouble num, char *bytes);
+void ConvertToIeeeDouble(  defdouble num, char *bytes);
+void ConvertToIeeeExtended(defdouble num, char *bytes);
+
+#endif

src/libespeak-ng/intonation.c

@@ -30,8 +30,8 @@
 
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
 #include "translate.h"
 
 /* Note this module is mostly old code that needs to be rewritten to
@@ -949,8 +949,8 @@ void CalcPitches(Translator *tr, int clause_type)
 	n_primary = 0;
 	for (ix = 0; ix < (n_phoneme_list-1); ix++) {
 		p = &phoneme_list[ix];
+		syllable_tab[ix].flags = 0;
 		if (p->synthflags & SFLAG_SYLLABLE) {
-			syllable_tab[n_st].flags = 0;
 			syllable_tab[n_st].env = PITCHfall;
 			syllable_tab[n_st].nextph_type = phoneme_list[ix+1].type;
 			syllable_tab[n_st++].stress = p->stresslevel;

src/libespeak-ng/klatt.c

@@ -35,10 +35,10 @@
 #include <espeak-ng/speak_lib.h>
 
 #include "speech.h"
-#include "klatt.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
+#include "klatt.h"
 
 extern unsigned char *out_ptr;
 extern unsigned char *out_start;
@@ -846,7 +846,7 @@ static int klattp[N_KLATTP];
 static double klattp1[N_KLATTP];
 static double klattp_inc[N_KLATTP];
 
-int Wavegen_Klatt(int resume)
+static int Wavegen_Klatt(int resume)
 {
 	int pk;
 	int x;
@@ -935,7 +935,7 @@ int Wavegen_Klatt(int resume)
 	return 0;
 }
 
-void SetSynth_Klatt(int length, frame_t *fr1, frame_t *fr2, voice_t *v, int control)
+static void SetSynth_Klatt(int length, frame_t *fr1, frame_t *fr2, voice_t *v, int control)
 {
 	int ix;
 	DOUBLEX next;

src/libespeak-ng/klatt.h

@@ -1,7 +1,7 @@
 /*
  * Copyright (C) 2008 by Jonathan Duddington
  * email: [email protected]
- * Copyright (C) 2015 Reece H. Dunn
+ * Copyright (C) 2015-2017 Reece H. Dunn
  *
  * Based on a re-implementation by:
  * (c) 1993,94 Jon Iles and Nick Ing-Simmons
@@ -171,6 +171,10 @@ typedef struct {
 	DOUBLEX bp_inc;
 }  klatt_peaks_t;
 
+void KlattInit(void);
+void KlattReset(int control);
+int Wavegen_Klatt2(int length, int resume, frame_t *fr1, frame_t *fr2);
+
 
 #ifdef __cplusplus
 }

src/libespeak-ng/mbrowrap.c

@@ -20,11 +20,24 @@
 #include "config.h"
 
 #if defined(_WIN32) || defined(_WIN64)
-
 #include <windows.h>
+#endif
 
 #include "mbrowrap.h"
 
+int (WINAPI *init_MBR)(char *voice_path);
+void (WINAPI *close_MBR)(void);
+void (WINAPI *reset_MBR)(void);
+int (WINAPI *read_MBR)(short *buffer, int nb_samples);
+int (WINAPI *write_MBR)(char *data);
+int (WINAPI *flush_MBR)(void);
+int (WINAPI *getFreq_MBR)(void);
+void (WINAPI *setVolumeRatio_MBR)(float value);
+char * (WINAPI *lastErrorStr_MBR)(char *buffer, int bufsize);
+void (WINAPI *setNoError_MBR)(int no_error);
+
+#if defined(_WIN32) || defined(_WIN64)
+
 HINSTANCE hinstDllMBR = NULL;
 
 BOOL load_MBR()
@@ -69,8 +82,6 @@ void unload_MBR()
 #include <unistd.h>
 
 #include <espeak-ng/espeak_ng.h>
-
-#include "mbrowrap.h"
 #include "speech.h"
 
 /*
@@ -508,7 +519,7 @@ static ssize_t receive_from_mbrola(void *buffer, size_t bufsize)
  * API functions.
  */
 
-int init_mbrola(char *voice_path)
+static int init_mbrola(char *voice_path)
 {
 	int error, result;
 	unsigned char wavhdr[45];
@@ -551,7 +562,7 @@ int init_mbrola(char *voice_path)
 	return 0;
 }
 
-void close_mbrola(void)
+static void close_mbrola(void)
 {
 	stop_mbrola();
 	free_pending_data();
@@ -560,7 +571,7 @@ void close_mbrola(void)
 	mbr_volume = 1.0;
 }
 
-void reset_mbrola(void)
+static void reset_mbrola(void)
 {
 	int result, success = 1;
 	char dummybuf[4096];
@@ -584,7 +595,7 @@ void reset_mbrola(void)
 		mbr_state = MBR_IDLE;
 }
 
-int read_mbrola(short *buffer, int nb_samples)
+static int read_mbrola(short *buffer, int nb_samples)
 {
 	int result = receive_from_mbrola(buffer, nb_samples * 2);
 	if (result > 0)
@@ -592,23 +603,23 @@ int read_mbrola(short *buffer, int nb_samples)
 	return result;
 }
 
-int write_mbrola(char *data)
+static int write_mbrola(char *data)
 {
 	mbr_state = MBR_NEWDATA;
 	return send_to_mbrola(data);
 }
 
-int flush_mbrola(void)
+static int flush_mbrola(void)
 {
 	return send_to_mbrola("\n#\n") == 3;
 }
 
-int getFreq_mbrola(void)
+static int getFreq_mbrola(void)
 {
 	return mbr_samplerate;
 }
 
-void setVolumeRatio_mbrola(float value)
+static void setVolumeRatio_mbrola(float value)
 {
 	if (value == mbr_volume)
 		return;
@@ -623,7 +634,7 @@ void setVolumeRatio_mbrola(float value)
 	init_MBR(mbr_voice_path);
 }
 
-char *lastErrorStr_mbrola(char *buffer, int bufsize)
+static char *lastErrorStr_mbrola(char *buffer, int bufsize)
 {
 	if (mbr_pid)
 		mbrola_has_errors();
@@ -631,7 +642,7 @@ char *lastErrorStr_mbrola(char *buffer, int bufsize)
 	return buffer;
 }
 
-void setNoError_mbrola(int no_error)
+static void setNoError_mbrola(int no_error)
 {
 	(void)no_error; // unused
 }

src/libespeak-ng/mbrowrap.h

@@ -37,20 +37,20 @@ typedef int BOOL;
  * error reason.  If this is successful, then close_MBR() must be called
  * before init_MBR() can be called again.
  */
-int (WINAPI *init_MBR)(char *voice_path);
+extern int (WINAPI *init_MBR)(char *voice_path);
 
 /*
  * Stop mbrola and release any resources.  It is necessary to call
  * this after a successful call to init_MBR() before init_MBR() can be
  * called again.
  */
-void (WINAPI *close_MBR)(void);
+extern void (WINAPI *close_MBR)(void);
 
 /*
  * Stop any ongoing processing and flush all buffers.  After this call
  * any synthesis request will start afresh.
  */
-void (WINAPI *reset_MBR)(void);
+extern void (WINAPI *reset_MBR)(void);
 
 /*
  * Return at most 'nb_samples' audio samples into 'buffer'. The returned
@@ -58,14 +58,14 @@ void (WINAPI *reset_MBR)(void);
  * If not successful, lastErrorStr_MBR() will provide the error reason.
  * Samples are always 16-bit little endian.
  */
-int (WINAPI *read_MBR)(short *buffer, int nb_samples);
+extern int (WINAPI *read_MBR)(short *buffer, int nb_samples);
 
 /*
  * Write a NULL terminated string of phoneme in the input buffer.
  * Return the number of chars actually written, or -1 on error.
  * If not successful, lastErrorStr_MBR() will provide the error reason.
  */
-int (WINAPI *write_MBR)(char *data);
+extern int (WINAPI *write_MBR)(char *data);
 
 /*
  * Send a flush command to the mbrola input stream.
@@ -73,17 +73,17 @@ int (WINAPI *write_MBR)(char *data);
  * or 0 on failure. If not successful, lastErrorStr_MBR() will provide
  * the error reason.
  */
-int (WINAPI *flush_MBR)(void);
+extern int (WINAPI *flush_MBR)(void);
 
 /*
  * Return the audio sample frequency of the used voice database.
  */
-int (WINAPI *getFreq_MBR)(void);
+extern int (WINAPI *getFreq_MBR)(void);
 
 /*
  * Overall volume.
  */
-void (WINAPI *setVolumeRatio_MBR)(float value);
+extern void (WINAPI *setVolumeRatio_MBR)(float value);
 
 /*
  * Copy into 'buffer' at most 'bufsize' bytes from the latest error
@@ -91,12 +91,12 @@ void (WINAPI *setVolumeRatio_MBR)(float value);
  * no error message is pending then an empty string is returned.
  * Consecutive calls to lastErrorStr_MBR() will return the same message.
  */
-char * (WINAPI *lastErrorStr_MBR)(char *buffer, int bufsize);
+extern char * (WINAPI *lastErrorStr_MBR)(char *buffer, int bufsize);
 
 /*
  * Tolerance to missing diphones.
  */
-void (WINAPI *setNoError_MBR)(int no_error);
+extern void (WINAPI *setNoError_MBR)(int no_error);
 
 BOOL load_MBR(void);
 void unload_MBR(void);

src/libespeak-ng/numbers.c

@@ -33,8 +33,8 @@
 
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
 #include "translate.h"
 
 #define M_LIGATURE  0x8000
@@ -548,7 +548,7 @@ static const int number_ranges[] = {
 	0
 };
 
-int NonAsciiNumber(int letter)
+static int NonAsciiNumber(int letter)
 {
 	// Change non-ascii digit into ascii digit '0' to '9', (or -1 if not)
 	const int *p;
@@ -1711,7 +1711,7 @@ static int LookupNum3(Translator *tr, int value, char *ph_out, int suppress_null
 	return 0;
 }
 
-bool CheckThousandsGroup(char *word, int group_len)
+static bool CheckThousandsGroup(char *word, int group_len)
 {
 	// Is this a group of 3 digits which looks like a thousands group?
 	int ix;

src/libespeak-ng/phonemelist.c

@@ -30,6 +30,7 @@
 
 #include "speech.h"
 #include "phoneme.h"
+#include "voice.h"
 #include "synthesize.h"
 #include "translate.h"
 

src/libespeak-ng/readclause.c

@@ -34,12 +34,13 @@
 #include <espeak-ng/espeak_ng.h>
 #include <espeak-ng/speak_lib.h>
 #include <espeak-ng/encoding.h>
+#include <ucd/ucd.h>
 
 #include "error.h"
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
 #include "translate.h"
 
 #define N_XML_BUF   500
@@ -89,6 +90,55 @@ PARAM_STACK param_stack[N_PARAM_STACK];
 static int speech_parameters[N_SPEECH_PARAM]; // current values, from param_stack
 int saved_parameters[N_SPEECH_PARAM]; // Parameters saved on synthesis start
 
+#define ESPEAKNG_CLAUSE_TYPE_PROPERTY_MASK 0xFFF0000000000000ull
+
+int clause_type_from_codepoint(uint32_t c)
+{
+	ucd_category cat = ucd_lookup_category(c);
+	ucd_property props = ucd_properties(c, cat);
+
+	switch (props & ESPEAKNG_CLAUSE_TYPE_PROPERTY_MASK)
+	{
+	case ESPEAKNG_PROPERTY_FULL_STOP:
+		return CLAUSE_PERIOD;
+	case ESPEAKNG_PROPERTY_FULL_STOP | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
+		return CLAUSE_PERIOD | CLAUSE_OPTIONAL_SPACE_AFTER;
+	case ESPEAKNG_PROPERTY_QUESTION_MARK:
+		return CLAUSE_QUESTION;
+	case ESPEAKNG_PROPERTY_QUESTION_MARK | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
+		return CLAUSE_QUESTION | CLAUSE_OPTIONAL_SPACE_AFTER;
+	case ESPEAKNG_PROPERTY_QUESTION_MARK | ESPEAKNG_PROPERTY_PUNCTUATION_IN_WORD:
+		return CLAUSE_QUESTION | CLAUSE_PUNCTUATION_IN_WORD;
+	case ESPEAKNG_PROPERTY_EXCLAMATION_MARK:
+		return CLAUSE_EXCLAMATION;
+	case ESPEAKNG_PROPERTY_EXCLAMATION_MARK | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
+		return CLAUSE_EXCLAMATION | CLAUSE_OPTIONAL_SPACE_AFTER;
+	case ESPEAKNG_PROPERTY_EXCLAMATION_MARK | ESPEAKNG_PROPERTY_PUNCTUATION_IN_WORD:
+		return CLAUSE_EXCLAMATION | CLAUSE_PUNCTUATION_IN_WORD;
+	case ESPEAKNG_PROPERTY_COMMA:
+		return CLAUSE_COMMA;
+	case ESPEAKNG_PROPERTY_COMMA | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
+		return CLAUSE_COMMA | CLAUSE_OPTIONAL_SPACE_AFTER;
+	case ESPEAKNG_PROPERTY_COLON:
+		return CLAUSE_COLON;
+	case ESPEAKNG_PROPERTY_COLON | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
+		return CLAUSE_COLON | CLAUSE_OPTIONAL_SPACE_AFTER;
+	case ESPEAKNG_PROPERTY_SEMI_COLON:
+	case ESPEAKNG_PROPERTY_EXTENDED_DASH:
+		return CLAUSE_SEMICOLON;
+	case ESPEAKNG_PROPERTY_SEMI_COLON | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
+	case ESPEAKNG_PROPERTY_QUESTION_MARK | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER | ESPEAKNG_PROPERTY_INVERTED_TERMINAL_PUNCTUATION:
+	case ESPEAKNG_PROPERTY_EXCLAMATION_MARK | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER | ESPEAKNG_PROPERTY_INVERTED_TERMINAL_PUNCTUATION:
+		return CLAUSE_SEMICOLON | CLAUSE_OPTIONAL_SPACE_AFTER;
+	case ESPEAKNG_PROPERTY_ELLIPSIS:
+		return CLAUSE_SEMICOLON | CLAUSE_SPEAK_PUNCTUATION_NAME | CLAUSE_OPTIONAL_SPACE_AFTER;
+	case ESPEAKNG_PROPERTY_PARAGRAPH_SEPARATOR:
+		return CLAUSE_PARAGRAPH;
+	}
+
+	return CLAUSE_NONE;
+}
+
 const int param_defaults[N_SPEECH_PARAM] = {
 	0,   // silence (internal use)
 	175, // rate wpm
@@ -113,7 +163,7 @@ int towlower2(unsigned int c)
 	if (c == 'I' && translator->langopts.dotless_i)
 		return 0x131; // I -> ı
 
-	return towlower(c);
+	return ucd_tolower(c);
 }
 
 static int IsRomanU(unsigned int c)
@@ -134,7 +184,6 @@ int Eof(void)
 static int GetC(void)
 {
 	int c1;
-	static int ungot2 = 0;
 
 	if ((c1 = ungot_char) != 0) {
 		ungot_char = 0;
@@ -792,7 +841,7 @@ static int attrnumber(const wchar_t *pw, int default_value, int type)
 
 	while (IsDigit09(*pw))
 		value = value*10 + *pw++ - '0';
-	if ((type == 1) && (towlower(*pw) == 's')) {
+	if ((type == 1) && (ucd_tolower(*pw) == 's')) {
 		// time: seconds rather than ms
 		value *= 1000;
 	}
@@ -869,7 +918,7 @@ static int attr_prosody_value(int param_type, const wchar_t *pw, int *value_out)
 	return sign;   // -1, 0, or 1
 }
 
-int AddNameData(const char *name, int wide)
+static int AddNameData(const char *name, int wide)
 {
 	// Add the name to the namedata and return its position
 	// (Used by the Windows SAPI wrapper)

src/libespeak-ng/setlengths.c

@@ -30,12 +30,10 @@
 
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
 #include "translate.h"
 
-extern int GetAmplitude(void);
-extern void DoSonicSpeed(int value);
 extern int saved_parameters[];
 
 // convert from words-per-minute to internal speed factor

src/libespeak-ng/spect.c

@@ -32,12 +32,10 @@
 
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
 #include "spect.h"
-
-extern double ConvertFromIeeeExtended(unsigned char *bytes);
-extern int PeaksToHarmspect(wavegen_peaks_t *peaks, int pitch, int *htab, int control);
+#include "ieee80.h"
 
 extern unsigned char pk_shape1[];
 extern int pk_select;
@@ -57,7 +55,7 @@ static double read_double(FILE *stream)
 {
 	unsigned char bytes[10];
 	fread(bytes, sizeof(char), 10, stream);
-	return ConvertFromIeeeExtended(bytes);
+	return ConvertFromIeeeExtended((char *)bytes);
 }
 
 float polint(float xa[], float ya[], int n, float x)

src/libespeak-ng/spect.h

@@ -22,6 +22,8 @@ extern "C"
 {
 #endif
 
+float polint(float xa[], float ya[], int n, float x);
+
 #define FRAME_WIDTH  1000  // max width for 8000kHz frame
 #define MAX_DISPLAY_FREQ 9500
 #define FRAME_HEIGHT 240
@@ -124,7 +126,7 @@ typedef struct {
 	int file_format;
 } SpectSeq;
 
-SpectSeq *SpectSeqCreate();
+SpectSeq *SpectSeqCreate(void);
 void SpectSeqDestroy(SpectSeq *spect);
 espeak_ng_STATUS LoadSpectSeq(SpectSeq *spect, const char *filename);
 

src/libespeak-ng/speech.c

@@ -50,8 +50,8 @@
 
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
 #include "translate.h"
 #include "espeak_command.h"
 #include "fifo.h"
@@ -400,7 +400,7 @@ static espeak_ng_STATUS Synthesize(unsigned int unique_identifier, const void *t
 
 	espeak_ng_STATUS status;
 	if (translator == NULL) {
-		status = espeak_SetVoiceByName("en");
+		status = espeak_ng_SetVoiceByName("en");
 		if (status != ENS_OK)
 			return status;
 	}
@@ -876,9 +876,6 @@ ESPEAK_NG_API espeak_ng_STATUS espeak_ng_Synchronize(void)
 	return berr;
 }
 
-extern void FreePhData(void);
-extern void FreeVoiceList(void);
-
 ESPEAK_NG_API espeak_ng_STATUS espeak_ng_Terminate(void)
 {
 #ifdef USE_ASYNC

src/libespeak-ng/synth_mbrola.c

@@ -33,14 +33,13 @@
 
 #include "speech.h"
 #include "phoneme.h"
+#include "voice.h"
 #include "synthesize.h"
 #include "translate.h"
-#include "voice.h"
 
 #ifdef INCLUDE_MBROLA
 
 extern int Read4Bytes(FILE *f);
-extern void SetPitch2(voice_t *voice, int pitch1, int pitch2, int *pitch_base, int *pitch_range);
 extern unsigned char *outbuf;
 
 #if defined(_WIN32) || defined(_WIN64)
@@ -573,16 +572,25 @@ void MbrolaReset(void)
 
 espeak_ng_STATUS LoadMbrolaTable(const char *mbrola_voice, const char *phtrans, int *srate)
 {
+	(void)mbrola_voice; // unused parameter
+	(void)phtrans; // unused parameter
+	(void)srate; // unused parameter
 	return ENS_NOT_SUPPORTED;
 }
 
 int MbrolaGenerate(PHONEME_LIST *phoneme_list, int *n_ph, int resume)
 {
+	(void)phoneme_list; // unused parameter
+	(void)n_ph; // unused parameter
+	(void)resume; // unused parameter
 	return 0;
 }
 
 int MbrolaFill(int length, int resume, int amplitude)
 {
+	(void)length; // unused parameter
+	(void)resume; // unused parameter
+	(void)amplitude; // unused parameter
 	return 0;
 }
 

src/libespeak-ng/synthdata.c

@@ -34,8 +34,8 @@
 #include "error.h"
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
 #include "translate.h"
 
 const char *version_string = PACKAGE_VERSION;
@@ -66,14 +66,12 @@ int vowel_transition[4];
 int vowel_transition0;
 int vowel_transition1;
 
-int FormantTransition2(frameref_t *seq, int *n_frames, unsigned int data1, unsigned int data2, PHONEME_TAB *other_ph, int which);
-
 static espeak_ng_STATUS ReadPhFile(void **ptr, const char *fname, int *size, espeak_ng_ERROR_CONTEXT *context)
 {
 	if (!ptr) return EINVAL;
 
 	FILE *f_in;
-	unsigned int length;
+	int length;
 	char buf[sizeof(path_home)+40];
 
 	sprintf(buf, "%s%c%s", path_home, PATHSEP, fname);
@@ -112,7 +110,6 @@ espeak_ng_STATUS LoadPhData(int *srate, espeak_ng_ERROR_CONTEXT *context)
 	int length = 0;
 	int rate;
 	unsigned char *p;
-	int *pw;
 
 	espeak_ng_STATUS status;
 	if ((status = ReadPhFile((void **)&phoneme_tab_data, "phontab", NULL, context)) != ENS_OK)
@@ -508,7 +505,6 @@ static bool InterpretCondition(Translator *tr, int control, PHONEME_LIST *plist,
 	unsigned int data;
 	int instn;
 	int instn2;
-	int count;
 	int check_endtype = 0;
 	PHONEME_TAB *ph;
 	PHONEME_LIST *plist_this;
@@ -760,7 +756,7 @@ int NumInstnWords(USHORT *prog)
 			// This instruction is followed by addWav(), 2 more words
 			return 4;
 		}
-		if (instn2 == OPCODE_CONTINUE)
+		if (instn2 == INSTN_CONTINUE)
 			return 3;
 		return 2;
 	}
@@ -817,10 +813,10 @@ void InterpretPhoneme(Translator *tr, int control, PHONEME_LIST *plist, PHONEME_
 				// instructions with no operand
 				switch (data)
 				{
-				case OPCODE_RETURN:
+				case INSTN_RETURN:
 					end_flag = 1;
 					break;
-				case OPCODE_CONTINUE:
+				case INSTN_CONTINUE:
 					break;
 				default:
 					InvalidInstn(ph, instn);
@@ -958,7 +954,7 @@ void InterpretPhoneme(Translator *tr, int control, PHONEME_LIST *plist, PHONEME_
 			param_sc = phdata->sound_param[instn2] = (instn >> 4) & 0xff;
 			prog++;
 
-			if (prog[1] != OPCODE_CONTINUE) {
+			if (prog[1] != INSTN_CONTINUE) {
 				if (instn2 < 2) {
 					// FMT() and WAV() imply Return
 					end_flag = 1;

src/libespeak-ng/synthesize.c

@@ -33,8 +33,8 @@
 
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
 #include "translate.h"
 
 extern FILE *f_log;

src/libespeak-ng/synthesize.h

@@ -74,7 +74,6 @@ extern "C"
 extern int embedded_value[N_EMBEDDED_VALUES];
 extern int embedded_default[N_EMBEDDED_VALUES];
 
-#define N_PEAKS   9
 #define N_PEAKS2  9 // plus Notch and Fill (not yet implemented)
 #define N_MARKERS 8
 
@@ -268,8 +267,8 @@ typedef struct {
 
 // instructions
 
-#define OPCODE_RETURN        0x0001
-#define OPCODE_CONTINUE      0x0002
+#define INSTN_RETURN         0x0001
+#define INSTN_CONTINUE       0x0002
 
 // Group 0 instrcutions with 8 bit operand.  These values go into bits 8-15 of the instruction
 #define i_CHANGE_PHONEME 0x01
@@ -453,14 +452,16 @@ extern int wcmdq_head;
 extern int wcmdq_tail;
 
 // from Wavegen file
-int  WcmdqFree();
-void WcmdqStop();
-int  WcmdqUsed();
-void WcmdqInc();
+int  WcmdqFree(void);
+void WcmdqStop(void);
+int  WcmdqUsed(void);
+void WcmdqInc(void);
 void WavegenInit(int rate, int wavemult_fact);
-float polint(float xa[], float ya[], int n, float x);
-int WavegenFill();
+int WavegenFill(void);
 void MarkerEvent(int type, unsigned int char_position, int value, int value2, unsigned char *out_ptr);
+int GetAmplitude(void);
+void SetPitch2(voice_t *voice, int pitch1, int pitch2, int *pitch_base, int *pitch_range);
+int PeaksToHarmspect(wavegen_peaks_t *peaks, int pitch, int *htab, int control);
 
 extern unsigned char *wavefile_data;
 extern int samplerate;
@@ -485,6 +486,7 @@ extern char mbrola_name[20];
 // from synthdata file
 unsigned int LookupSound(PHONEME_TAB *ph1, PHONEME_TAB *ph2, int which, int *match_level, int control);
 frameref_t *LookupSpect(PHONEME_TAB *this_ph, int which, FMT_PARAMS *fmt_params,  int *n_frames, PHONEME_LIST *plist);
+void FreePhData(void);
 
 unsigned char *LookupEnvelope(int ix);
 espeak_ng_STATUS LoadPhData(int *srate, espeak_ng_ERROR_CONTEXT *context);
@@ -497,11 +499,16 @@ void SetSpeed(int control);
 void SetEmbedded(int control, int value);
 void SelectPhonemeTable(int number);
 int  SelectPhonemeTableName(const char *name);
+int FormantTransition2(frameref_t *seq, int *n_frames, unsigned int data1, unsigned int data2, PHONEME_TAB *other_ph, int which);
 
 void Write4Bytes(FILE *f, int value);
 int Read4Bytes(FILE *f);
 int Reverse4Bytes(int word);
 
+#if HAVE_SONIC_H
+void DoSonicSpeed(int value);
+#endif
+
 #define ENV_LEN  128    // length of pitch envelopes
 #define PITCHfall   0  // standard pitch envelopes
 #define PITCHrise   2
@@ -545,10 +552,6 @@ int NumInstnWords(USHORT *prog);
 
 void InitBreath(void);
 
-void KlattInit();
-void KlattReset(int control);
-int Wavegen_Klatt2(int length, int resume, frame_t *fr1, frame_t *fr2);
-
 #ifdef __cplusplus
 }
 #endif

src/libespeak-ng/tokenizer.c

@@ -1,410 +0,0 @@
-/* Tokenizer APIs.
- *
- * Copyright (C) 2017 Reece H. Dunn
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 3 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, see: <http://www.gnu.org/licenses/>.
- */
-
-#include "config.h"
-
-#include <errno.h>
-#include <stdint.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <string.h>
-
-#include <espeak-ng/espeak_ng.h>
-#include <espeak-ng/encoding.h>
-#include <espeak-ng/tokenizer.h>
-#include <ucd/ucd.h>
-
-#include "speech.h"
-#include "phoneme.h"
-#include "synthesize.h"
-#include "translate.h"
-
-#define ESPEAKNG_CLAUSE_TYPE_PROPERTY_MASK 0xFFF0000000000000ull
-
-int clause_type_from_codepoint(uint32_t c)
-{
-	ucd_category cat = ucd_lookup_category(c);
-	ucd_property props = ucd_properties(c, cat);
-
-	switch (props & ESPEAKNG_CLAUSE_TYPE_PROPERTY_MASK)
-	{
-	case ESPEAKNG_PROPERTY_FULL_STOP:
-		return CLAUSE_PERIOD;
-	case ESPEAKNG_PROPERTY_FULL_STOP | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
-		return CLAUSE_PERIOD | CLAUSE_OPTIONAL_SPACE_AFTER;
-	case ESPEAKNG_PROPERTY_QUESTION_MARK:
-		return CLAUSE_QUESTION;
-	case ESPEAKNG_PROPERTY_QUESTION_MARK | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
-		return CLAUSE_QUESTION | CLAUSE_OPTIONAL_SPACE_AFTER;
-	case ESPEAKNG_PROPERTY_QUESTION_MARK | ESPEAKNG_PROPERTY_PUNCTUATION_IN_WORD:
-		return CLAUSE_QUESTION | CLAUSE_PUNCTUATION_IN_WORD;
-	case ESPEAKNG_PROPERTY_EXCLAMATION_MARK:
-		return CLAUSE_EXCLAMATION;
-	case ESPEAKNG_PROPERTY_EXCLAMATION_MARK | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
-		return CLAUSE_EXCLAMATION | CLAUSE_OPTIONAL_SPACE_AFTER;
-	case ESPEAKNG_PROPERTY_EXCLAMATION_MARK | ESPEAKNG_PROPERTY_PUNCTUATION_IN_WORD:
-		return CLAUSE_EXCLAMATION | CLAUSE_PUNCTUATION_IN_WORD;
-	case ESPEAKNG_PROPERTY_COMMA:
-		return CLAUSE_COMMA;
-	case ESPEAKNG_PROPERTY_COMMA | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
-		return CLAUSE_COMMA | CLAUSE_OPTIONAL_SPACE_AFTER;
-	case ESPEAKNG_PROPERTY_COLON:
-		return CLAUSE_COLON;
-	case ESPEAKNG_PROPERTY_COLON | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
-		return CLAUSE_COLON | CLAUSE_OPTIONAL_SPACE_AFTER;
-	case ESPEAKNG_PROPERTY_SEMI_COLON:
-	case ESPEAKNG_PROPERTY_EXTENDED_DASH:
-		return CLAUSE_SEMICOLON;
-	case ESPEAKNG_PROPERTY_SEMI_COLON | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER:
-	case ESPEAKNG_PROPERTY_QUESTION_MARK | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER | ESPEAKNG_PROPERTY_INVERTED_TERMINAL_PUNCTUATION:
-	case ESPEAKNG_PROPERTY_EXCLAMATION_MARK | ESPEAKNG_PROPERTY_OPTIONAL_SPACE_AFTER | ESPEAKNG_PROPERTY_INVERTED_TERMINAL_PUNCTUATION:
-		return CLAUSE_SEMICOLON | CLAUSE_OPTIONAL_SPACE_AFTER;
-	case ESPEAKNG_PROPERTY_ELLIPSIS:
-		return CLAUSE_SEMICOLON | CLAUSE_SPEAK_PUNCTUATION_NAME | CLAUSE_OPTIONAL_SPACE_AFTER;
-	case ESPEAKNG_PROPERTY_PARAGRAPH_SEPARATOR:
-		return CLAUSE_PARAGRAPH;
-	}
-
-	return CLAUSE_NONE;
-}
-
-typedef enum {
-	ESPEAKNG_CTYPE_OTHER,
-	ESPEAKNG_CTYPE_CARRIAGE_RETURN,
-	ESPEAKNG_CTYPE_NEWLINE,
-	ESPEAKNG_CTYPE_END_OF_STRING,
-	ESPEAKNG_CTYPE_PARAGRAPH,
-	ESPEAKNG_CTYPE_WHITESPACE,
-	ESPEAKNG_CTYPE_LOWERCASE,
-	ESPEAKNG_CTYPE_UPPERCASE,
-	ESPEAKNG_CTYPE_FULL_STOP,
-	ESPEAKNG_CTYPE_QUESTION_MARK,
-	ESPEAKNG_CTYPE_EXCLAMATION_MARK,
-	ESPEAKNG_CTYPE_COMMA,
-	ESPEAKNG_CTYPE_COLON,
-	ESPEAKNG_CTYPE_SEMICOLON,
-	ESPEAKNG_CTYPE_ELLIPSIS,
-	ESPEAKNG_CTYPE_PUNCTUATION,
-	ESPEAKNG_CTYPE_SYMBOL,
-} espeakng_CTYPE;
-
-#define ESPEAKNG_CTYPE_PROPERTY_MASK 0xFE0000000000C001ull
-
-// Reference: http://www.unicode.org/reports/tr14/tr14-32.html -- Unicode Line Breaking Algorithm
-static espeakng_CTYPE codepoint_type(uint32_t c)
-{
-	// 1. Detect and classify specific codepoints.
-
-	switch (c)
-	{
-	case 0x0000: return ESPEAKNG_CTYPE_END_OF_STRING; // NULL
-	case 0x000A: return ESPEAKNG_CTYPE_NEWLINE; // LINE FEED (LF)
-	case 0x000C: return ESPEAKNG_CTYPE_NEWLINE; // FORM FEED (FF)
-	case 0x000D: return ESPEAKNG_CTYPE_CARRIAGE_RETURN; // CARRIAGE RETURN (CR)
-	case 0x0085: return ESPEAKNG_CTYPE_NEWLINE; // NEW LINE (NEL)
-	}
-
-	// 2. Override property types for codepoints by their Unicode General Category.
-
-	ucd_category cat = ucd_lookup_category(c);
-	switch (cat)
-	{
-	case UCD_CATEGORY_Zl: return ESPEAKNG_CTYPE_NEWLINE;
-	case UCD_CATEGORY_Zp: return ESPEAKNG_CTYPE_PARAGRAPH;
-	}
-
-	// 3. Classify codepoints by their Unicode properties.
-
-	ucd_property props = ucd_properties(c, cat);
-	switch (props & ESPEAKNG_CTYPE_PROPERTY_MASK)
-	{
-	case UCD_PROPERTY_WHITE_SPACE:
-		return ESPEAKNG_CTYPE_WHITESPACE;
-	case UCD_PROPERTY_OTHER_LOWERCASE:
-		return ESPEAKNG_CTYPE_LOWERCASE;
-	case UCD_PROPERTY_OTHER_UPPERCASE:
-		return ESPEAKNG_CTYPE_UPPERCASE;
-	case ESPEAKNG_PROPERTY_FULL_STOP:
-		return ESPEAKNG_CTYPE_FULL_STOP;
-	case ESPEAKNG_PROPERTY_QUESTION_MARK:
-		return ESPEAKNG_CTYPE_QUESTION_MARK;
-	case ESPEAKNG_PROPERTY_EXCLAMATION_MARK:
-		return ESPEAKNG_CTYPE_EXCLAMATION_MARK;
-	case ESPEAKNG_PROPERTY_COMMA:
-		return ESPEAKNG_CTYPE_COMMA;
-	case ESPEAKNG_PROPERTY_COLON:
-		return ESPEAKNG_CTYPE_COLON;
-	case ESPEAKNG_PROPERTY_SEMI_COLON:
-		return ESPEAKNG_CTYPE_SEMICOLON;
-	case ESPEAKNG_PROPERTY_ELLIPSIS:
-		return ESPEAKNG_CTYPE_ELLIPSIS;
-	}
-
-	// 4. Classify codepoints by their Unicode General Category.
-
-	switch (cat)
-	{
-	case UCD_CATEGORY_Lu: return ESPEAKNG_CTYPE_UPPERCASE;
-	case UCD_CATEGORY_Ll: return ESPEAKNG_CTYPE_LOWERCASE;
-	case UCD_CATEGORY_Pc: return ESPEAKNG_CTYPE_PUNCTUATION;
-	case UCD_CATEGORY_Pd: return ESPEAKNG_CTYPE_PUNCTUATION;
-	case UCD_CATEGORY_Pe: return ESPEAKNG_CTYPE_PUNCTUATION;
-	case UCD_CATEGORY_Pf: return ESPEAKNG_CTYPE_PUNCTUATION;
-	case UCD_CATEGORY_Pi: return ESPEAKNG_CTYPE_PUNCTUATION;
-	case UCD_CATEGORY_Po: return ESPEAKNG_CTYPE_PUNCTUATION;
-	case UCD_CATEGORY_Ps: return ESPEAKNG_CTYPE_PUNCTUATION;
-	case UCD_CATEGORY_Sc: return ESPEAKNG_CTYPE_SYMBOL;
-	case UCD_CATEGORY_Sk: return ESPEAKNG_CTYPE_SYMBOL;
-	case UCD_CATEGORY_Sm: return ESPEAKNG_CTYPE_SYMBOL;
-	case UCD_CATEGORY_So: return ESPEAKNG_CTYPE_SYMBOL;
-	}
-
-	// 5. Classify the remaining codepoints.
-
-	return ESPEAKNG_CTYPE_OTHER;
-}
-
-#define ESPEAKNG_CODEPOINT_INVALID 0xFFFFFFFF
-
-struct espeak_ng_TOKENIZER_
-{
-	espeak_ng_TEXT_DECODER *decoder;
-	char token[256];
-	uint32_t keepc;
-
-	espeak_ng_TOKEN_TYPE (*read)(espeak_ng_TOKENIZER *tokenizer);
-};
-
-static espeak_ng_TOKEN_TYPE
-tokenizer_state_end_of_buffer(espeak_ng_TOKENIZER *tokenizer)
-{
-	*tokenizer->token = '\0';
-	return ESPEAKNG_TOKEN_END_OF_BUFFER;
-}
-
-static espeak_ng_TOKEN_TYPE
-tokenizer_read_word_token(espeak_ng_TOKENIZER *tokenizer, char *current, espeak_ng_TOKEN_TYPE type)
-{
-	char *end = tokenizer->token + sizeof(tokenizer->token) - 5; // allow for UTF-8 trailing bytes
-	int initial_state = 1;
-
-	while (current < end && !text_decoder_eof(tokenizer->decoder)) {
-		uint32_t c = text_decoder_getc(tokenizer->decoder);
-		switch (codepoint_type(c))
-		{
-		case ESPEAKNG_CTYPE_LOWERCASE:
-			current += utf8_out(c, current);
-			switch (type)
-			{
-			case ESPEAKNG_TOKEN_WORD_LOWERCASE:
-			case ESPEAKNG_TOKEN_WORD_MIXEDCASE:
-			case ESPEAKNG_TOKEN_WORD_CAPITALIZED:
-				break;
-			case ESPEAKNG_TOKEN_WORD_UPPERCASE:
-				type = initial_state
-				     ? ESPEAKNG_TOKEN_WORD_CAPITALIZED
-				     : ESPEAKNG_TOKEN_WORD_MIXEDCASE;
-				break;
-			}
-			initial_state = 0;
-			break;
-		case ESPEAKNG_CTYPE_UPPERCASE:
-			current += utf8_out(c, current);
-			switch (type)
-			{
-			case ESPEAKNG_TOKEN_WORD_UPPERCASE:
-			case ESPEAKNG_TOKEN_WORD_MIXEDCASE:
-				break;
-			case ESPEAKNG_TOKEN_WORD_LOWERCASE:
-			case ESPEAKNG_TOKEN_WORD_CAPITALIZED:
-				type = ESPEAKNG_TOKEN_WORD_MIXEDCASE;
-				break;
-			}
-			initial_state = 0;
-			break;
-		default:
-			tokenizer->keepc = c;
-			*current = '\0';
-			return type;
-		}
-	}
-
-	*current = '\0';
-	return type;
-}
-
-static espeak_ng_TOKEN_TYPE
-tokenizer_state_default(espeak_ng_TOKENIZER *tokenizer)
-{
-	if (text_decoder_eof(tokenizer->decoder)) {
-		tokenizer->read = tokenizer_state_end_of_buffer;
-		return tokenizer_state_end_of_buffer(tokenizer);
-	}
-
-	char *current = tokenizer->token;
-	char *end     = tokenizer->token + sizeof(tokenizer->token) - 5; // allow for UTF-8 trailing bytes
-
-	uint32_t c;
-	if (tokenizer->keepc != ESPEAKNG_CODEPOINT_INVALID) {
-		c = tokenizer->keepc;
-		tokenizer->keepc = ESPEAKNG_CODEPOINT_INVALID;
-	} else {
-		c = text_decoder_getc(tokenizer->decoder);
-	}
-
-	switch (codepoint_type(c))
-	{
-	case ESPEAKNG_CTYPE_END_OF_STRING: // '\0'
-		tokenizer->read = tokenizer_state_end_of_buffer;
-		return tokenizer_state_end_of_buffer(tokenizer);
-	case ESPEAKNG_CTYPE_CARRIAGE_RETURN: // '\r'
-		if (text_decoder_peekc(tokenizer->decoder) == '\n') {
-			current += utf8_out(c, current);
-			c = text_decoder_getc(tokenizer->decoder);
-		}
-		// fallthrough
-	case ESPEAKNG_CTYPE_NEWLINE:
-		current += utf8_out(c, current);
-		*current = '\0';
-		return ESPEAKNG_TOKEN_NEWLINE;
-	case ESPEAKNG_CTYPE_PARAGRAPH:
-		current += utf8_out(c, current);
-		*current = '\0';
-		return ESPEAKNG_TOKEN_PARAGRAPH;
-	case ESPEAKNG_CTYPE_WHITESPACE:
-		current += utf8_out(c, current);
-		while (!text_decoder_eof(tokenizer->decoder) &&
-		       current < end &&
-		       codepoint_type(c = text_decoder_getc(tokenizer->decoder)) == ESPEAKNG_CTYPE_WHITESPACE)
-		{
-			current += utf8_out(c, current);
-		}
-		tokenizer->keepc = c;
-		*current = '\0';
-		return ESPEAKNG_TOKEN_WHITESPACE;
-	case ESPEAKNG_CTYPE_LOWERCASE:
-		current += utf8_out(c, current);
-		return tokenizer_read_word_token(tokenizer, current, ESPEAKNG_TOKEN_WORD_LOWERCASE);
-	case ESPEAKNG_CTYPE_UPPERCASE:
-		current += utf8_out(c, current);
-		return tokenizer_read_word_token(tokenizer, current, ESPEAKNG_TOKEN_WORD_UPPERCASE);
-	case ESPEAKNG_CTYPE_FULL_STOP:
-		current += utf8_out(c, current);
-		if (c == '.' && text_decoder_peekc(tokenizer->decoder) == '.') {
-			c = text_decoder_getc(tokenizer->decoder);
-			if (text_decoder_peekc(tokenizer->decoder) == '.') {
-				c = text_decoder_getc(tokenizer->decoder);
-				current += utf8_out('.', current);
-				current += utf8_out('.', current);
-				*current = '\0';
-				return ESPEAKNG_TOKEN_ELLIPSIS;
-			} else {
-				tokenizer->keepc = c;
-			}
-		}
-		*current = '\0';
-		return ESPEAKNG_TOKEN_FULL_STOP;
-	case ESPEAKNG_CTYPE_QUESTION_MARK:
-		current += utf8_out(c, current);
-		*current = '\0';
-		return ESPEAKNG_TOKEN_QUESTION_MARK;
-	case ESPEAKNG_CTYPE_EXCLAMATION_MARK:
-		current += utf8_out(c, current);
-		*current = '\0';
-		return ESPEAKNG_TOKEN_EXCLAMATION_MARK;
-	case ESPEAKNG_CTYPE_COMMA:
-		current += utf8_out(c, current);
-		*current = '\0';
-		return ESPEAKNG_TOKEN_COMMA;
-	case ESPEAKNG_CTYPE_COLON:
-		current += utf8_out(c, current);
-		*current = '\0';
-		return ESPEAKNG_TOKEN_COLON;
-	case ESPEAKNG_CTYPE_SEMICOLON:
-		current += utf8_out(c, current);
-		*current = '\0';
-		return ESPEAKNG_TOKEN_SEMICOLON;
-	case ESPEAKNG_CTYPE_ELLIPSIS:
-		current += utf8_out(c, current);
-		*current = '\0';
-		return ESPEAKNG_TOKEN_ELLIPSIS;
-	case ESPEAKNG_CTYPE_PUNCTUATION:
-		current += utf8_out(c, current);
-		*current = '\0';
-		return ESPEAKNG_TOKEN_PUNCTUATION;
-	case ESPEAKNG_CTYPE_SYMBOL:
-		current += utf8_out(c, current);
-		*current = '\0';
-		return ESPEAKNG_TOKEN_SYMBOL;
-	default:
-		current += utf8_out(c, current);
-		*current = '\0';
-		return ESPEAKNG_TOKEN_UNKNOWN;
-	}
-
-	return ESPEAKNG_TOKEN_END_OF_BUFFER;
-}
-
-#pragma GCC visibility push(default)
-
-espeak_ng_TOKENIZER *
-create_tokenizer(void)
-{
-	espeak_ng_TOKENIZER *tokenizer = malloc(sizeof(espeak_ng_TOKENIZER));
-	if (!tokenizer) return NULL;
-
-	tokenizer->decoder = NULL;
-	tokenizer->keepc = ESPEAKNG_CODEPOINT_INVALID;
-	tokenizer->read = tokenizer_state_end_of_buffer;
-
-	*tokenizer->token = '\0';
-	return tokenizer;
-}
-
-void
-destroy_tokenizer(espeak_ng_TOKENIZER *tokenizer)
-{
-	if (tokenizer) free(tokenizer);
-}
-
-int
-tokenizer_reset(espeak_ng_TOKENIZER *tokenizer,
-                espeak_ng_TEXT_DECODER *decoder,
-                espeak_ng_TOKENIZER_OPTIONS options)
-{
-	if (!tokenizer) return 0;
-
-	tokenizer->decoder = decoder;
-	tokenizer->keepc = ESPEAKNG_CODEPOINT_INVALID;
-	tokenizer->read = decoder ? tokenizer_state_default : tokenizer_state_end_of_buffer;
-	return 1;
-}
-
-espeak_ng_TOKEN_TYPE
-tokenizer_read_next_token(espeak_ng_TOKENIZER *tokenizer)
-{
-	return tokenizer->read(tokenizer);
-}
-
-const char *
-tokenizer_get_token_text(espeak_ng_TOKENIZER *tokenizer)
-{
-	return tokenizer->token;
-}
-
-#pragma GCC visibility pop

src/libespeak-ng/tr_languages.c

@@ -32,6 +32,7 @@
 
 #include "speech.h"
 #include "phoneme.h"
+#include "voice.h"
 #include "synthesize.h"
 #include "translate.h"
 
@@ -158,19 +159,44 @@ static void SetLetterBitsRange(Translator *tr, int group, int first, int last)
 
 // ignore these characters
 static const unsigned short chars_ignore_default[] = {
-	0xad,    1, // soft hyphen
-	0x200c,  1, // zero width non-joiner
-	0x200d,  1, // zero width joiner
-	0, 0
+	// U+00AD SOFT HYPHEN
+	//     Used to mark hyphenation points in words for where to split a
+	//     word at the end of a line to provide readable justified text.
+	0xad,   1,
+	// U+200C ZERO WIDTH NON-JOINER
+	//     Used to prevent combined ligatures being displayed in their
+	//     combined form.
+	0x200c, 1,
+	// U+200D ZERO WIDTH JOINER
+	//     Used to indicate an alternative connected form made up of the
+	//     characters surrounding the ZWJ in Devanagari, Kannada, Malayalam
+	//     and Emoji.
+//	0x200d, 1, // Not ignored.
+	// End of the ignored character list.
+	0,      0
 };
 
 // alternatively, ignore characters but allow zero-width-non-joiner (lang-fa)
 static const unsigned short chars_ignore_zwnj_hyphen[] = {
-	0xad,   1,   // soft hyphen
-	0x640,  1,   // igniore Arabic Tatweel (lang=FA)
-	0x200c, '-', // zero width non-joiner, replace with hyphen
-	0x200d, 1,   // zero width joiner
-	0, 0
+	// U+00AD SOFT HYPHEN
+	//     Used to mark hyphenation points in words for where to split a
+	//     word at the end of a line to provide readable justified text.
+	0xad,   1,
+	// U+0640 TATWEEL (KASHIDA)
+	//     Used in Arabic scripts to stretch characters for justifying
+	//     the text.
+	0x640,  1,
+	// U+200C ZERO WIDTH NON-JOINER
+	//     Used to prevent combined ligatures being displayed in their
+	//     combined form.
+	0x200c, '-',
+	// U+200D ZERO WIDTH JOINER
+	//     Used to indicate an alternative connected form made up of the
+	//     characters surrounding the ZWJ in Devanagari, Kannada, Malayalam
+	//     and Emoji.
+//	0x200d, 1, // Not ignored.
+	// End of the ignored character list.
+	0,      0
 };
 
 const unsigned char utf8_ordinal[] = { 0xc2, 0xba, 0 }; // masculine ordinal character, UTF-8
@@ -387,7 +413,7 @@ static void SetCyrillicLetters(Translator *tr)
 	SetLetterBits(tr, LETTERGP_VOWEL2, (char *)ru_vowels);
 }
 
-void SetIndicLetters(Translator *tr)
+static void SetIndicLetters(Translator *tr)
 {
 	// Set letter types for Indic scripts, Devanagari, Tamill, etc
 	static const char dev_consonants2[] = { 0x02, 0x03, 0x58, 0x59, 0x5a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f, 0x7b, 0x7c, 0x7e, 0x7f, 0 };
@@ -412,7 +438,7 @@ void SetIndicLetters(Translator *tr)
 	tr->langopts.suffix_add_e = tr->letter_bits_offset + 0x4d; // virama
 }
 
-void SetupTranslator(Translator *tr, const short *lengths, const unsigned char *amps)
+static void SetupTranslator(Translator *tr, const short *lengths, const unsigned char *amps)
 {
 	if (lengths != NULL)
 		memcpy(tr->stress_lengths, lengths, sizeof(tr->stress_lengths));

src/libespeak-ng/translate.c

@@ -33,8 +33,8 @@
 
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
 #include "translate.h"
 
 Translator *translator = NULL; // the main translator
@@ -254,7 +254,7 @@ int IsDigit(unsigned int c)
 	return 0;
 }
 
-int IsSpace(unsigned int c)
+static int IsSpace(unsigned int c)
 {
 	if (c == 0)
 		return 0;
@@ -265,6 +265,16 @@ int IsSpace(unsigned int c)
 	return iswspace(c);
 }
 
+int isspace2(unsigned int c)
+{
+	// can't use isspace() because on Windows, isspace(0xe1) gives TRUE !
+	int c2;
+
+	if (((c2 = (c & 0xff)) == 0) || (c > ' '))
+		return 0;
+	return 1;
+}
+
 void DeleteTranslator(Translator *tr)
 {
 	if (tr->data_dictlist != NULL)
@@ -388,6 +398,40 @@ int utf8_in(int *c, const char *buf)
 }
 #pragma GCC visibility pop
 
+int utf8_out(unsigned int c, char *buf)
+{
+	// write a unicode character into a buffer as utf8
+	// returns the number of bytes written
+
+	int n_bytes;
+	int j;
+	int shift;
+	static char unsigned code[4] = { 0, 0xc0, 0xe0, 0xf0 };
+
+	if (c < 0x80) {
+		buf[0] = c;
+		return 1;
+	}
+	if (c >= 0x110000) {
+		buf[0] = ' '; // out of range character code
+		return 1;
+	}
+	if (c < 0x0800)
+		n_bytes = 1;
+	else if (c < 0x10000)
+		n_bytes = 2;
+	else
+		n_bytes = 3;
+
+	shift = 6*n_bytes;
+	buf[0] = code[n_bytes] | (c >> shift);
+	for (j = 0; j < n_bytes; j++) {
+		shift -= 6;
+		buf[j+1] = 0x80 + ((c >> shift) & 0x3f);
+	}
+	return n_bytes+1;
+}
+
 char *strchr_w(const char *s, int c)
 {
 	// return NULL for any non-ascii character
@@ -396,17 +440,6 @@ char *strchr_w(const char *s, int c)
 	return strchr((char *)s, c); // (char *) is needed for Borland compiler
 }
 
-int IsAllUpper(const char *word)
-{
-	int c;
-	while ((*word != 0) && !isspace2(*word)) {
-		word += utf8_in(&c, word);
-		if (!iswupper(c))
-			return 0;
-	}
-	return 1;
-}
-
 static char *SpeakIndividualLetters(Translator *tr, char *word, char *phonemes, int spell_word)
 {
 	int posn = 0;
@@ -525,7 +558,6 @@ static int TranslateWord3(Translator *tr, char *word_start, WORD_TAB *wtab, char
 	int more_suffixes;
 	int confirm_prefix;
 	int spell_word;
-	int stress_bits;
 	int emphasize_allcaps = 0;
 	int wflags;
 	int wmark;
@@ -1139,7 +1171,7 @@ int TranslateWord(Translator *tr, char *word_start, WORD_TAB *wtab, char *word_o
 		char word[N_WORD_BYTES+1];
 		word[0] = 0;
 		word[1] = ' ';
-		memcpy(word+2, word_out, strlen(word_out));
+		strcpy(word+2, word_out);
 		word_out = word+2;
 
 		while (*word_out && available > 1) {
@@ -1195,7 +1227,7 @@ static int CountSyllables(unsigned char *phonemes)
 	return count;
 }
 
-void Word_EmbeddedCmd()
+static void Word_EmbeddedCmd()
 {
 	// Process embedded commands for emphasis, sayas, and break
 	int embedded_cmd;
@@ -1791,13 +1823,13 @@ static int SubstituteChar(Translator *tr, unsigned int c, unsigned int next_in,
 		// don't convert the case of the second character unless the next letter is also upper case
 		c2 = new_c >> 16;
 		if (upper_case && iswupper(next_in))
-			c2 = towupper(c2);
+			c2 = ucd_toupper(c2);
 		*insert = c2;
 		new_c &= 0xffff;
 	}
 
 	if (upper_case)
-		new_c = towupper(new_c);
+		new_c = ucd_toupper(new_c);
 
 	*wordflags |= FLAG_CHAR_REPLACED;
 	return new_c;
@@ -1874,7 +1906,7 @@ static int TranslateChar(Translator *tr, char *ptr, int prev_in, unsigned int c,
 
 static const char *UCase_ga[] = { "bp", "bhf", "dt", "gc", "hA", "mb", "nd", "ng", "ts", "tA", "nA", NULL };
 
-int UpperCaseInWord(Translator *tr, char *word, int c)
+static int UpperCaseInWord(Translator *tr, char *word, int c)
 {
 	int ix;
 	int len;

src/libespeak-ng/translate.h

@@ -1,7 +1,7 @@
 /*
  * Copyright (C) 2005 to 2014 by Jonathan Duddington
  * email: [email protected]
- * Copyright (C) 2015 Reece H. Dunn
+ * Copyright (C) 2015-2017 Reece H. Dunn
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -706,10 +706,15 @@ extern void SetLengthMods(Translator *tr, int value);
 
 void LoadConfig(void);
 int TransposeAlphabet(Translator *tr, char *text);
+
+#define LEADING_2_BITS 0xC0 // 0b11000000
+#define UTF8_TAIL_BITS 0x80 // 0b10000000
+
 ESPEAK_NG_API int utf8_in(int *c, const char *buf);
 int utf8_in2(int *c, const char *buf, int backwards);
 int utf8_out(unsigned int c, char *buf);
 int utf8_nbytes(const char *buf);
+
 int lookupwchar(const unsigned short *list, int c);
 int lookupwchar2(const unsigned short *list, int c);
 int Eof(void);
@@ -747,6 +752,10 @@ void LookupAccentedLetter(Translator *tr, unsigned int letter, char *ph_buf);
 
 int LoadDictionary(Translator *tr, const char *name, int no_error);
 int LookupDictList(Translator *tr, char **wordptr, char *ph_out, unsigned int *flags, int end_flags, WORD_TAB *wtab);
+int HashDictionary(const char *string);
+
+void print_dictionary_flags(unsigned int *flags, char *buf, int buf_len);
+char *DecodeRule(const char *group_chars, int group_length, char *rule, int control);
 
 void MakePhonemeList(Translator *tr, int post_pause, int new_sentence);
 int ChangePhonemes_ru(Translator *tr, PHONEME_LIST2 *phlist, int n_ph, int index, PHONEME_TAB *ph, CHANGEPH *ch);

src/libespeak-ng/voice.h

@@ -22,6 +22,8 @@ extern "C"
 {
 #endif
 
+#define N_PEAKS   9
+
 typedef struct {
 	char v_name[40];
 	char language_name[20];
@@ -86,6 +88,7 @@ espeak_ng_STATUS DoVoiceChange(voice_t *v);
 void WavegenSetVoice(voice_t *v);
 void ReadTonePoints(char *string, int *tone_pts);
 void VoiceReset(int control);
+void FreeVoiceList(void);
 
 #ifdef __cplusplus
 }

src/libespeak-ng/voices.c

@@ -40,8 +40,8 @@
 
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
 #include "translate.h"
 
 MNEM_TAB genders[] = {
@@ -474,18 +474,6 @@ static int Read8Numbers(char *data_in, int *data)
 	              &data[0], &data[1], &data[2], &data[3], &data[4], &data[5], &data[6], &data[7]);
 }
 
-static unsigned int StringToWord2(const char *string)
-{
-	// Convert a language name string to a word such as L('e','n')
-	int ix;
-	int c;
-	unsigned int value = 0;
-
-	for (ix = 0; (ix < 4) && ((c = string[ix]) != 0); ix++)
-		value = (value << 8) | (c & 0xff);
-	return value;
-}
-
 voice_t *LoadVoice(const char *vname, int control)
 {
 	// control, bit 0  1= no_default
@@ -544,7 +532,7 @@ voice_t *LoadVoice(const char *vname, int control)
 			return NULL;
 	} else {
 		if (voicename[0] == 0)
-			strcpy(voicename, "en");
+			strcpy(voicename, ESPEAKNG_DEFAULT_VOICE);
 
 		sprintf(path_voices, "%s%cvoices%c", path_home, PATHSEP, PATHSEP);
 		sprintf(buf, "%s%s", path_voices, voicename); // look in the main voices directory
@@ -705,47 +693,57 @@ voice_t *LoadVoice(const char *vname, int control)
 			break;
 		case V_INTONATION: // intonation
 			sscanf(p, "%d %d", &option_tone_flags, &option_tone2);
-			if ((option_tone_flags & 0xff) != 0)
-				langopts->intonation_group = option_tone_flags & 0xff;
+			if ((option_tone_flags & 0xff) != 0) {
+				if (langopts)
+					langopts->intonation_group = option_tone_flags & 0xff;
+				else
+					fprintf(stderr, "Cannot set intonation: language not set, or is invalid.\n");
+			}
 			break;
 		case V_TUNES:
 			n = sscanf(p, "%s %s %s %s %s %s", names[0], names[1], names[2], names[3], names[4], names[5]);
-			langopts->intonation_group = 0;
-			for (ix = 0; ix < n; ix++) {
-				if (strcmp(names[ix], "NULL") == 0)
-					continue;
-
-				if ((value = LookupTune(names[ix])) < 0)
-					fprintf(stderr, "Unknown tune '%s'\n", names[ix]);
-				else
-					langopts->tunes[ix] = value;
-			}
+			if (langopts) {
+				langopts->intonation_group = 0;
+				for (ix = 0; ix < n; ix++) {
+					if (strcmp(names[ix], "NULL") == 0)
+						continue;
+
+					if ((value = LookupTune(names[ix])) < 0)
+						fprintf(stderr, "Unknown tune '%s'\n", names[ix]);
+					else
+						langopts->tunes[ix] = value;
+				}
+			} else
+				fprintf(stderr, "Cannot set tunes: language not set, or is invalid.\n");
 			break;
 		case V_DICTRULES: // conditional dictionary rules and list entries
 		case V_NUMBERS:
 		case V_STRESSOPT:
-			// expect a list of numbers
-			while (*p != 0) {
-				while (isspace(*p)) p++;
-				if ((n = atoi(p)) > 0) {
-					p++;
-					if (n < 32) {
-						if (key == V_DICTRULES)
-							conditional_rules |= (1 << n);
-						else if (key == V_NUMBERS)
-							langopts->numbers |= (1 << n);
-						else if (key == V_STRESSOPT)
-							langopts->stress_flags |= (1 << n);
-					} else {
-						if ((key == V_NUMBERS) && (n < 64))
-							langopts->numbers2 |= (1 << (n-32));
-						else
-							fprintf(stderr, "Bad option number %d\n", n);
+			if (langopts) {
+				// expect a list of numbers
+				while (*p != 0) {
+					while (isspace(*p)) p++;
+					if ((n = atoi(p)) > 0) {
+						p++;
+						if (n < 32) {
+							if (key == V_DICTRULES)
+								conditional_rules |= (1 << n);
+							else if (key == V_NUMBERS)
+								langopts->numbers |= (1 << n);
+							else if (key == V_STRESSOPT)
+								langopts->stress_flags |= (1 << n);
+						} else {
+							if ((key == V_NUMBERS) && (n < 64))
+								langopts->numbers2 |= (1 << (n-32));
+							else
+								fprintf(stderr, "Bad option number %d\n", n);
+						}
 					}
+					while (isalnum(*p)) p++;
 				}
-				while (isalnum(*p)) p++;
-			}
-			ProcessLanguageOptions(langopts);
+				ProcessLanguageOptions(langopts);
+			} else
+				fprintf(stderr, "Cannot set stressopt: language not set, or is invalid.\n");
 			break;
 		case V_REPLACE:
 			if (phonemes_set == 0) {
@@ -756,22 +754,31 @@ voice_t *LoadVoice(const char *vname, int control)
 			PhonemeReplacement(p);
 			break;
 		case V_WORDGAP: // words
-			sscanf(p, "%d %d", &langopts->word_gap, &langopts->vowel_pause);
+			if (langopts)
+				sscanf(p, "%d %d", &langopts->word_gap, &langopts->vowel_pause);
+			else
+				fprintf(stderr, "Cannot set wordgap: language not set, or is invalid.\n");
 			break;
 		case V_STRESSRULE:
-			sscanf(p, "%d %d %d %d", &langopts->stress_rule,
-			       &langopts->stress_flags,
-			       &langopts->unstressed_wd1,
-			       &langopts->unstressed_wd2);
+			if (langopts)
+				sscanf(p, "%d %d %d %d", &langopts->stress_rule,
+				       &langopts->stress_flags,
+				       &langopts->unstressed_wd1,
+				       &langopts->unstressed_wd2);
+			else
+				fprintf(stderr, "Cannot set stressrule: language not set, or is invalid.\n");
 			break;
 		case V_OPTION:
-			value2 = 0;
-			if (((sscanf(p, "%s %d %d", option_name, &value, &value2) >= 2) && ((ix = LookupMnem(options_tab, option_name)) >= 0)) ||
-			    ((sscanf(p, "%d %d %d", &ix, &value, &value2) >= 2) && (ix < N_LOPTS))) {
-				langopts->param[ix] = value;
-				langopts->param2[ix] = value2;
+			if (langopts) {
+				value2 = 0;
+				if (((sscanf(p, "%s %d %d", option_name, &value, &value2) >= 2) && ((ix = LookupMnem(options_tab, option_name)) >= 0)) ||
+				    ((sscanf(p, "%d %d %d", &ix, &value, &value2) >= 2) && (ix < N_LOPTS))) {
+					langopts->param[ix] = value;
+					langopts->param2[ix] = value2;
+				} else
+					fprintf(stderr, "Bad voice option: %s %s\n", buf, p);
 			} else
-				fprintf(stderr, "Bad voice option: %s %s\n", buf, p);
+				fprintf(stderr, "Cannot set option: language not set, or is invalid.\n");
 			break;
 		case V_ECHO:
 			// echo.  suggest: 135mS  11%
@@ -852,9 +859,12 @@ voice_t *LoadVoice(const char *vname, int control)
 		case V_STATUS:
 			break;
 		default:
-			if ((key & 0xff00) == 0x100)
-				sscanf(p, "%d", &langopts->param[key &0xff]);
-			else
+			if ((key & 0xff00) == 0x100) {
+				if (langopts)
+					sscanf(p, "%d", &langopts->param[key &0xff]);
+				else
+					fprintf(stderr, "Cannot set voice attribute: language not set, or is invalid.\n");
+			} else
 				fprintf(stderr, "Bad voice attribute: %s\n", buf);
 			break;
 		}
@@ -1272,7 +1282,7 @@ char const *SelectVoice(espeak_VOICE *voice_select, int *found)
 
 		if (voice_select2.name == NULL) {
 			if ((voice_select2.name = voice_select2.identifier) == NULL)
-				voice_select2.name = "default";
+				voice_select2.name = ESPEAKNG_DEFAULT_VOICE;
 		}
 
 		strncpy0(buf, voice_select2.name, sizeof(buf));
@@ -1299,7 +1309,7 @@ char const *SelectVoice(espeak_VOICE *voice_select, int *found)
 	if (nv == 0) {
 		// no matching voice, choose the default
 		*found = 0;
-		if ((voices[0] = SelectVoiceByName(voices_list, "default")) != NULL)
+		if ((voices[0] = SelectVoiceByName(voices_list, ESPEAKNG_DEFAULT_VOICE)) != NULL)
 			nv = 1;
 	}
 

src/libespeak-ng/wavegen.c

@@ -33,8 +33,12 @@
 
 #include "speech.h"
 #include "phoneme.h"
-#include "synthesize.h"
 #include "voice.h"
+#include "synthesize.h"
+
+#ifdef INCLUDE_KLATT
+#include "klatt.h"
+#endif
 
 #if HAVE_SONIC_H
 #include "sonic.h"
@@ -665,7 +669,7 @@ static int ApplyBreath(void)
 	return value;
 }
 
-int Wavegen()
+static int Wavegen()
 {
 	if (wvoice == NULL)
 		return 0;
@@ -1117,7 +1121,7 @@ void SetPitch2(voice_t *voice, int pitch1, int pitch2, int *pitch_base, int *pit
 	*pitch_range = base + (pitch2 * range)/2 - *pitch_base;
 }
 
-void SetPitch(int length, unsigned char *env, int pitch1, int pitch2)
+static void SetPitch(int length, unsigned char *env, int pitch1, int pitch2)
 {
 	if (wvoice == NULL)
 		return;
@@ -1140,7 +1144,7 @@ void SetPitch(int length, unsigned char *env, int pitch1, int pitch2)
 	flutter_amp = wvoice->flutter;
 }
 
-void SetSynth(int length, int modn, frame_t *fr1, frame_t *fr2, voice_t *v)
+static void SetSynth(int length, int modn, frame_t *fr1, frame_t *fr2, voice_t *v)
 {
 	if (wvoice == NULL || v == NULL)
 		return;
@@ -1246,7 +1250,7 @@ void Write4Bytes(FILE *f, int value)
 	}
 }
 
-int WavegenFill2()
+static int WavegenFill2()
 {
 	// Pick up next wavegen commands from the queue
 	// return: 0  output buffer has been filled

src/ucd-tools/.gitignore

@@ -1,4 +1,5 @@
 .*.swp
+*~
 
 # intermediate files:
 
@@ -52,12 +53,15 @@ depcomp
 install-sh
 libtool
 ltmain.sh
-m4/
 missing
 Makefile
 Makefile.in
 stamp-h1
 
+m4/*
+!m4/ax_check_compile_flag.m4
+
 # emscripten:
+
 a.out.js
 a.out.js.mem

src/ucd-tools/CHANGELOG.md

@@ -7,11 +7,14 @@ These are eSpeak NG specific modifications to the `ucd-tools` project:
 *  `data/espeak-ng` data files for eSpeak NG extended data.
 *  espeak-ng PropList property lookup as part of the `ucd_property` API.
 
-## 9.0.0.1 - (In Progress)
+## 10.0.0 - 2017-06-25
 
 *  Add `iswblank` and `iswxdigit` compatibility.
 *  Improve ctype compatibility.
-*  PropList property lookup.
+*  PropList and emoji-data property lookup.
+*  Support building with a C89 compiler.
+*  Update to Unicode Character Data 10.0.0.
+*  Unicode Emoji 5.0.
 
 ## 9.0.0 - 2016-12-28
 

src/ucd-tools/Makefile.am

@@ -55,34 +55,34 @@ EXTRA_DIST += ChangeLog
 
 ############################# Unicode Data ####################################
 
-EMOJI_VERSION=4.0
+EMOJI_VERSION=5.0
 UCD_VERSION=@UCD_VERSION@
 UCD_ROOTDIR=data/ucd
 UCD_SRCDIR=http://www.unicode.org/Public
 
 data/emoji/emoji-data.txt:
 	mkdir -pv data/emoji
-	curl ${UCD_SRCDIR}/emoji/${EMOJI_VERSION}/emoji-data.txt > $@
+	curl ${UCD_SRCDIR}/emoji/${EMOJI_VERSION}/emoji-data.txt -o $@
 
 data/ucd/PropList.txt:
 	mkdir -pv data/ucd
-	curl ${UCD_SRCDIR}/${UCD_VERSION}/ucd/PropList.txt > $@
+	curl ${UCD_SRCDIR}/${UCD_VERSION}/ucd/PropList.txt -o $@
 
 data/ucd/DerivedCoreProperties.txt:
 	mkdir -pv data/ucd
-	curl ${UCD_SRCDIR}/${UCD_VERSION}/ucd/DerivedCoreProperties.txt > $@
+	curl ${UCD_SRCDIR}/${UCD_VERSION}/ucd/DerivedCoreProperties.txt -o $@
 
 data/ucd/PropertyValueAliases.txt:
 	mkdir -pv data/ucd
-	curl ${UCD_SRCDIR}/${UCD_VERSION}/ucd/PropertyValueAliases.txt > $@
+	curl ${UCD_SRCDIR}/${UCD_VERSION}/ucd/PropertyValueAliases.txt -o $@
 
 data/ucd/Scripts.txt:
 	mkdir -pv data/ucd
-	curl ${UCD_SRCDIR}/${UCD_VERSION}/ucd/Scripts.txt > $@
+	curl ${UCD_SRCDIR}/${UCD_VERSION}/ucd/Scripts.txt -o $@
 
 data/ucd/UnicodeData.txt:
 	mkdir -pv data/ucd
-	curl ${UCD_SRCDIR}/${UCD_VERSION}/ucd/UnicodeData.txt > $@
+	curl ${UCD_SRCDIR}/${UCD_VERSION}/ucd/UnicodeData.txt -o $@
 
 ############################# documentation ###################################
 

src/ucd-tools/configure.ac

@@ -1,5 +1,5 @@
 AC_PREREQ([2.65])
-AC_INIT([Unicode Character Database Tools], [9.0.0], [https://github.com/rhdunn/ucd-tools/issues], [ucd-tools], [https://github.com/rhdunn/ucd-tools])
+AC_INIT([Unicode Character Database Tools], [10.0.0], [https://github.com/rhdunn/ucd-tools/issues], [ucd-tools], [https://github.com/rhdunn/ucd-tools])
 AM_INIT_AUTOMAKE()
 
 m4_ifdef([AM_SILENT_RULES],[AM_SILENT_RULES])
@@ -24,6 +24,7 @@ dnl library checks.
 dnl ================================================================
 
 AC_CHECK_HEADERS([stddef.h])   dnl C89
+AC_CHECK_FUNCS([iswblank])     dnl C99
 
 AC_TYPE_UINT8_T
 AC_TYPE_UINT32_T
@@ -33,13 +34,22 @@ dnl UCD data configuration.
 dnl ================================================================
 
 AC_ARG_WITH([unicode-version],
-    [AS_HELP_STRING([--with-unicode-version], [Unicode version to support @<:@default=9.0.0@:>@])],
+    [AS_HELP_STRING([--with-unicode-version], [Unicode version to support @<:@default=10.0.0@:>@])],
     [AS_IF([test x"$withval" != x],
            [UCD_VERSION="$withval"])],
-    [UCD_VERSION="9.0.0"])
+    [UCD_VERSION="10.0.0"])
 
 AC_SUBST(UCD_VERSION)
 
+dnl ================================================================
+dnl Compiler warnings.
+dnl
+dnl Reference: https://gcc.gnu.org/onlinedocs/gcc/Warning-Options.html
+dnl ================================================================
+
+AX_CHECK_COMPILE_FLAG([-Wmissing-prototypes],    [CFLAGS="-Wmissing-prototypes $CFLAGS"])
+AX_CHECK_COMPILE_FLAG([-Wreturn-type],           [CFLAGS="-Wreturn-type $CFLAGS"])
+
 dnl ================================================================
 dnl Generate output.
 dnl ================================================================

src/ucd-tools/m4/ax_check_compile_flag.m4

@@ -0,0 +1,75 @@
+# ===========================================================================
+#  https://www.gnu.org/software/autoconf-archive/ax_check_compile_flag.html
+# ===========================================================================
+#
+# SYNOPSIS
+#
+#   AX_CHECK_COMPILE_FLAG(FLAG, [ACTION-SUCCESS], [ACTION-FAILURE], [EXTRA-FLAGS], [INPUT])
+#
+# DESCRIPTION
+#
+#   Check whether the given FLAG works with the current language's compiler
+#   or gives an error.  (Warnings, however, are ignored)
+#
+#   ACTION-SUCCESS/ACTION-FAILURE are shell commands to execute on
+#   success/failure.
+#
+#   If EXTRA-FLAGS is defined, it is added to the current language's default
+#   flags (e.g. CFLAGS) when the check is done.  The check is thus made with
+#   the flags: "CFLAGS EXTRA-FLAGS FLAG".  This can for example be used to
+#   force the compiler to issue an error when a bad flag is given.
+#
+#   INPUT gives an alternative input source to AC_COMPILE_IFELSE.
+#
+#   NOTE: Implementation based on AX_CFLAGS_GCC_OPTION. Please keep this
+#   macro in sync with AX_CHECK_{PREPROC,LINK}_FLAG.
+#
+# LICENSE
+#
+#   Copyright (c) 2008 Guido U. Draheim <[email protected]>
+#   Copyright (c) 2011 Maarten Bosmans <[email protected]>
+#
+#   This program is free software: you can redistribute it and/or modify it
+#   under the terms of the GNU General Public License as published by the
+#   Free Software Foundation, either version 3 of the License, or (at your
+#   option) any later version.
+#
+#   This program is distributed in the hope that it will be useful, but
+#   WITHOUT ANY WARRANTY; without even the implied warranty of
+#   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General
+#   Public License for more details.
+#
+#   You should have received a copy of the GNU General Public License along
+#   with this program. If not, see <https://www.gnu.org/licenses/>.
+#
+#   As a special exception, the respective Autoconf Macro's copyright owner
+#   gives unlimited permission to copy, distribute and modify the configure
+#   scripts that are the output of Autoconf when processing the Macro. You
+#   need not follow the terms of the GNU General Public License when using
+#   or distributing such scripts, even though portions of the text of the
+#   Macro appear in them. The GNU General Public License (GPL) does govern
+#   all other use of the material that constitutes the Autoconf Macro.
+#
+#   This special exception to the GPL applies to versions of the Autoconf
+#   Macro released by the Autoconf Archive. When you make and distribute a
+#   modified version of the Autoconf Macro, you may extend this special
+#   exception to the GPL to apply to your modified version as well.
+
+#serial 5
+
+AC_DEFUN([AX_CHECK_COMPILE_FLAG],
+[AC_PREREQ(2.64)dnl for _AC_LANG_PREFIX and AS_VAR_IF
+AS_VAR_PUSHDEF([CACHEVAR],[ax_cv_check_[]_AC_LANG_ABBREV[]flags_$4_$1])dnl
+AC_CACHE_CHECK([whether _AC_LANG compiler accepts $1], CACHEVAR, [
+  ax_check_save_flags=$[]_AC_LANG_PREFIX[]FLAGS
+  _AC_LANG_PREFIX[]FLAGS="$[]_AC_LANG_PREFIX[]FLAGS $4 $1"
+  AC_COMPILE_IFELSE([m4_default([$5],[AC_LANG_PROGRAM()])],
+    [AS_VAR_SET(CACHEVAR,[yes])],
+    [AS_VAR_SET(CACHEVAR,[no])])
+  _AC_LANG_PREFIX[]FLAGS=$ax_check_save_flags])
+AS_VAR_IF(CACHEVAR,yes,
+  [m4_default([$2], :)],
+  [m4_default([$3], :)])
+AS_VAR_POPDEF([CACHEVAR])dnl
+])dnl AX_CHECK_COMPILE_FLAGS
+

src/ucd-tools/src/case.c

@@ -18,14 +18,15 @@
  * along with ucd-tools.  If not, see <http://www.gnu.org/licenses/>.
  */
 
-// NOTE: This file is automatically generated from the UnicodeData.txt file in
-// the Unicode Character database by the ucd-tools/tools/categories.py script.
+/* NOTE: This file is automatically generated from the UnicodeData.txt file in
+ * the Unicode Character database by the ucd-tools/tools/categories.py script.
+ */
 
 #include "ucd/ucd.h"
 
 #include <stddef.h>
 
-// Unicode Character Data 9.0.0
+/* Unicode Character Data 10.0.0 */
 
 struct case_conversion_entry
 {