[1.47.08]

Add API function: espeak_TextToPhonemes().


git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@349 d46cf337-b52f-0410-862d-fd96e6ae7743

dictsource/dict_phonemes

 ts   v    x    z    
 
 
-Dictionary en_dict  2013-04-25
+Dictionary en_dict  2013-04-26
 
 0    0#   3    3:   @    @-   @2   @5   
 @L   a    a#   A:   A@   aa   aI   aI3  
 ts;  v    w    x    z    Z    Z;   
 
 
-Dictionary pt_dict  2013-04-25
+Dictionary pt_dict  2013-04-26
 
 &    &/   &U~  &~   @    @-   a    A    
 aI   aU   e    E    eI   EI   eU   EU   

dictsource/en_list

 _¡	%Inv,3:tI2dEkskl@m'eIS@n
 _¦	broUk@nb'A@
 _¨	daIa#r'i:sIs
-_ª	fEmI2nIn'O@dI2n@L
 _«	lEft'gIl@mEt
 _­	makr0n
-²	tu:
-³	Tri:
-_¹	su:p@skrIptw'0n
-_²	su:p@skrIptt'u:
-_³	su:p@skrIptT'ri:
+²	tu:	$native
+³	Tri:	$native
+
 _´	a#kju:taksEnt
 _·	mId@Ld0t
 _¸	sEd'Il@
-//_º	maskjUlIn'O@dI2n@L
-_º	dI2gr'i:z
 _»	raIt'gIl@mEt
 _¿	%Inv,3:tI2dkw'EstS@n
 
 foyer		fOIeI
 freelance	fri:laans
 freon		fri:0n
-frequent	frI2kwEnt $verb
 fricassee	frIka#seI
 frigate		frIg@t
 fringer		frIndZ3	// infringer
 Andre		0ndreI
 Andrea		andri@
 Andreas		A:ndr'eI@s
-Angela		$1
+Angela		ang@l@
 Anita		$alt3
 Anthony		ant@ni
 Anton		ant0n

dictsource/en_rules

             ery (A       3j
         _v) ery          ErI2
         qu) ery          i@rI
+        _x) er (o        i@r
          z) er (o        i@r
         _C) er (o_       i@r
             er (rA       E

dictsource/pt_list

 doce		$alt2
 doutora		$alt2
 doze		$alt2
-droga		$alt
 dueto		$alt2
 elixir		eliSir
 emprego		$alt2 $noun
 rota		$alt $noun
 rubores		$alt2
 rumores		$alt2
-sabonete		$alt2
+sabonete	$alt2
 sabores		$alt2
 saibro		s'aIbRU
 saleta		$alt2
 ?1 line       l'ain
    layout     l'eI,aUt
    ?2   login  logiN
-?2        logo        l'OgU
 ?1 lock       l'Ok
    mail       m'eIl
         mails        m'eIls

dictsource/pt_rules

 
      i) e (r_      E
     @s) e (r_      E
-     qu) e (r_      E
+    qu) e (r_      E
 
     ib) e (rL04_   E
      d) e (rL04_   E
      f) e (rL02_   E
 
         e (quL02_  E
-e (rbL04_   E
+        e (rbL04_  E
         e (rgL04_  E
 
         e (rsL04_  E
         e (ssL04_  E
         e (ss_     E         // stress and english words
 ?2      ection (_  'EkS&~N
-?2   ee (n_   i
+?2      ee (n_     i
         e (stL04_  E
 
         e (strL04_ E
    tiv) e (L05_    E         // es-
    viv) e (L05_    e
   _com) e (L05_    e
-  s_colh) e (res_   E
   _cos) e (L05_    e
   _pud) e (L05_    E
  _coub) e (L05_    E
 _interromp) e (L07_ e
 _troux) e (L07_    E
   _ens) e (bL03_   E
-  _enver) e (dL03_   E
 _amanc) e (bL03_   E
     qu) e (brL04_  E
   _cel) e (brL04_  E
  _boch) e (chL04_  e
- _on) e (rL04_   E
-    _ex) e (crL04_  E
+   _ex) e (crL04_  E
   _enr) e (dL01_   E
     _v) e (dL03_   E
   _arr) e (dL03_   E
  _depr) e (dL03_   E
  _hosp) e (dL03_   E
+_enver) e (dL03_   E
    _gr) e (gL03_   e
   _gal) e (gL03_   e
   _pel) e (gL03_   e
     _g) e (rL04_   E
     _z) e (rL04_   E
     qu) e (rL04_   E         // re-
+   _on) e (rL04_   E
    _op) e (rL04_   E
    gen) e (rL04_   E         // de- re-
   _alt) e (rL04_   E
 _persev) e (rL04_  E
 _recup) e (rL04_   E
 _refrig) e (rL04_  E
-       _ex) e (rcL02_  E
+   _ex) e (rcL02_  E
  _alic) e (rcL02_  E
     _p) e (rdL02_  E
     _l) e (rdL03_  E
     _h) e (rdL04_  E
   _des) e (rdL04_  E
+s_colh) e (res_    E
      _) e (rgL03_  e
-_so) e (rgL03_   e 
    _re) e (rgL03_  e
+   _so) e (rgL03_  e
   _bez) e (rrL03_  e
      v) e (rtL03_  e         // con- re- in-
     _f) e (rvL03_  e
    óCC) o (CA_     o
     úC) o (CA_     o
    úCC) o (CA_     o
-?2   otion (_   'owS&~N
-   
+?2      otion (_   'owS&~N
+
 //sort
     _f) o (L05_    o
 _pormen) o (L05_   O
   _esn) o (bL04_   O
  _engl) o (bL04_   O
   _sal) o (brL03_  o
-  _env) o (ltL03_   o
      _) o (cL03_   o
     _m) o (fL01_   O
     _m) o (fL02_   O
     _s) o (frL02_  O
+     l) o (gL04_   O         // dialoga, etc.
      r) o (gL04_   O         // inter-
+     v) o (gL04_   O         // advoga, etc.
     _j) o (gL04_   O
     af) o (gL04_   O         // des-
-  _adv) o (gL04_   O
   _ref) o (gL04_   O
    _ap) oi (L01_   OI
    _ap) oi (L02_   OI
   _emp) o (lgL04_  O
      _) o (lhL04_  O
      m) o (lhL04_  O
+  _env) o (ltL03_  o
    ent) o (pL02_   O
     _c) o (pL03_   O
     _t) o (pL04_   O
         u (a       u
      g) u (a+      w
      q) u (a+      w
-u (eC    u
-u (ei    u
+        u (eC      u
+        u (ei      u
      r) u (_A      u
         u (L04_    'u
 
         ui (_      'uI
         u (iu      w
         u (iCK     u
-u (éC   u
-u (íC   u
+        u (éC      u
+        u (íC      u
         uy         uI
         ul (K      uw
         u (lh+     u
 
      g) u (Y 
     _m) ui (tA_    'wiN
-?2   ution (_   'uS&~N
+?2      ution (_   'uS&~N
 
 
 .group ú

platforms/windows/make_espeak.iss

 
 [Setup]
 AppName=eSpeak
-AppVerName=eSpeak version 1.47.06
+AppVerName=eSpeak version 1.47.08
 AppCopyright=Licensed under GNU General Public License version 3.   (See file License.txt for details).
 WindowVisible=yes
 

platforms/windows/make_espeakedit.iss

 
 [Setup]
 AppName=eSpeakEdit
-AppVerName=eSpeakEdit version 1.47.06
+AppVerName=eSpeakEdit version 1.47.08
 DefaultDirName={pf}\eSpeak
 DefaultGroupName=eSpeak
 OutputBaseFilename=setup_espeakedit

src/compiledict.cpp

 	{"$stem",    0x30},   // must have a suffix
 	{"$atend",   0x31},   // use this pronunciation if at end of clause
 	{"$atstart", 0x32},   // use this pronunciation at start of clause
+	{"$native",  0x33},   // not if we've switched translators
 
 	// doesn't set dictionary_flags
 	{"$?",        100},   // conditional rule, followed by byte giving the condition number

src/dictionary.cpp

 				continue;
 			}
 		}
+		if(dictionary_flags2 & FLAG_NATIVE)
+		{
+			if(tr != translator)
+				continue;    // don't use if we've switched translators
+		}
 		if(dictionary_flags & FLAG_ALT2_TRANS)
 		{
 			// language specific

src/espeakedit.cpp

    EVT_MENU(MENU_OPT_SPEED, MyFrame::OnOptions)
    EVT_MENU(MENU_OPT_PUNCT, MyFrame::OnOptions)
    EVT_MENU(MENU_OPT_SPELL, MyFrame::OnOptions)
+	EVT_MENU(MENU_OPT_SPELL2, MyFrame::OnOptions)
    EVT_MENU(MENU_PATH0, MyFrame::OnOptions)
    EVT_MENU(MENU_PATH1, MyFrame::OnOptions)
    EVT_MENU(MENU_PATH2, MyFrame::OnOptions)
 		break;
 
 	case MENU_OPT_SPELL:
+		transldlg->t_source->SetValue(_T("<say-as interpret-as=\"characters\">\n"));
+		transldlg->t_source->SetInsertionPointEnd();
+		notebook->SetSelection(1);
+		break;
+
+	case MENU_OPT_SPELL2:
 		transldlg->t_source->SetValue(_T("<say-as interpret-as=\"tts:char\">\n"));
 		transldlg->t_source->SetInsertionPointEnd();
 		notebook->SetSelection(1);

src/main.h

 	MENU_OPT_SPEED,
 	MENU_OPT_PUNCT,
 	MENU_OPT_SPELL,
+	MENU_OPT_SPELL2,
 	MENU_PATHS,
 	MENU_PATH0,
 	MENU_PATH1,

src/menus.cpp

         option_menu = new wxMenu;
         option_menu->Append(MENU_PATHS, _("Set &paths"), paths_menu);
         option_menu->Append(MENU_OPT_SPEED, _("Speed..."));
+        option_menu->AppendSeparator();
         option_menu->Append(MENU_OPT_PUNCT, _("Speak punctuation"));
-        option_menu->Append(MENU_OPT_SPELL, _("Speak character names"));
+        option_menu->Append(MENU_OPT_SPELL, _("Speak characters"));
+        option_menu->Append(MENU_OPT_SPELL2, _("Speak character name"));
 
         // TOOLS
         vowelchart_menu = new wxMenu;

src/numbers.cpp

 		}
 	}
 	letter = towlower2(letter);
+	LookupLetter(tr, letter, word[n_bytes], ph_buf, control & 1);
 
-	// is this a subscript or superscript letter ?
-	for(ix=0; (c = derived_letters[ix]) != 0; ix+=2)
+	if(ph_buf[0] == 0)
 	{
-		if(c > letter)
-			break;
-		if(c == letter)
+		// is this a subscript or superscript letter ?
+		for(ix=0; (c = derived_letters[ix]) != 0; ix+=2)
 		{
-			c = derived_letters[ix+1];
-			letter = c & 0x3fff;
-			if((modifier = modifiers[c >> 14]) != NULL)
+			if(c > letter)
+				break;
+			if(c == letter)
 			{
-				Lookup(tr, modifier, capital);
-				if(capital[0] == 0)
+				c = derived_letters[ix+1];
+				letter = c & 0x3fff;
+				if((modifier = modifiers[c >> 14]) != NULL)
 				{
-					capital[2] = SetTranslator2("en");   // overwrites previous contents of translator2
-					Lookup(translator2, modifier, &capital[3]);
-					if(capital[3] != 0)
+					Lookup(tr, modifier, capital);
+					if(capital[0] == 0)
 					{
-						capital[0] = phonPAUSE;
-						capital[1] = phonSWITCH;
-						len = strlen(&capital[3]);
-						capital[len+3] = phonSWITCH;
-						capital[len+4] = phontab_1;
-						capital[len+5] = 0;
+						capital[2] = SetTranslator2("en");   // overwrites previous contents of translator2
+						Lookup(translator2, modifier, &capital[3]);
+						if(capital[3] != 0)
+						{
+							capital[0] = phonPAUSE;
+							capital[1] = phonSWITCH;
+							len = strlen(&capital[3]);
+							capital[len+3] = phonSWITCH;
+							capital[len+4] = phontab_1;
+							capital[len+5] = 0;
+						}
 					}
 				}
 			}
 		}
+		LookupLetter(tr, letter, word[n_bytes], ph_buf, control & 1);
 	}
-	LookupLetter(tr, letter, word[n_bytes], ph_buf, control & 1);
 
 	if(ph_buf[0] == phonSWITCH)
 	{

src/readclause.cpp

 		return(towupper(c));
 
 	// check whether a previous character code is the upper-case equivalent of this character
-	if(towlower2(c-32) == c)
+	if(towlower2(c-32) == (int)c)
 		return(c-32); // yes, use it
-	if(towlower2(c-1) == c)
+	if(towlower2(c-1) == (int)c)
 		return(c-1);
 	for(ix=0; wchar_toupper[ix] != 0; ix+=2)
 	{
-		if(wchar_toupper[ix] == c)
+		if(wchar_toupper[ix] == (int)c)
 			return(wchar_toupper[ix+1]);
 	}
 	return(c);  // no

src/speak_lib.cpp

 ESPEAK_API void espeak_TextToPhonemes(const void *text, char *buffer, int size, int textmode, int phonememode)
 {//==============================================================================================================
 	/* phoneme_mode  bits 0-3: 0=only phoneme names, 1=ties, 2=ZWJ, 3=underscore separator
-	                 bit  4:   0=eSpeak phoneme names, 1=IPA
+	                 bits 4-7:   0=eSpeak phoneme names, 1=IPA
 	*/
 	option_multibyte = textmode & 7;
 	TranslateClause(translator, NULL, text, NULL, NULL);

src/speak_lib.h

   Changed espeak_EVENT structure to add id.string[] for phoneme mnemonics.
   Added espeakINITIALIZE_PHONEME_IPA option for espeak_Initialize() to report phonemes as IPA names.
 
+Revision 8  26.Apr.2013
+  Added function espeak_TextToPhonemes().
+
 */
          /********************/
          /*  Initialization  */
 		char string[8];    // used for phoneme names (UTF8). Terminated by a zero byte unless the name needs the full 8 bytes.
 	} id;
 } espeak_EVENT;
-/* 
+/*
    When a message is supplied to espeak_synth, the request is buffered and espeak_synth returns. When the message is really processed, the callback function will be repetedly called.
 
 
 
    In PLAYBACK mode, the callback function is called as soon as an event happens.
 
-   For example suppose that the following message is supplied to espeak_Synth: 
+   For example suppose that the following message is supplied to espeak_Synth:
    "hello, hello."
 
 
    * Once processed in RETRIEVAL mode, it could lead to 3 calls of the callback function :
 
    ** Block 1:
-   <audio data> + 
+   <audio data> +
    List of events: SENTENCE + WORD + LIST_TERMINATED
- 
+
    ** Block 2:
    <audio data> +
    List of events: WORD + END + LIST_TERMINATED
 
 typedef enum {
 	/* PLAYBACK mode: plays the audio data, supplies events to the calling program*/
-	AUDIO_OUTPUT_PLAYBACK, 
+	AUDIO_OUTPUT_PLAYBACK,
 
 	/* RETRIEVAL mode: supplies audio data and events to the calling program */
 	AUDIO_OUTPUT_RETRIEVAL,
- 
+
 	/* SYNCHRONOUS mode: as RETRIEVAL but doesn't return until synthesis is completed */
 	AUDIO_OUTPUT_SYNCHRONOUS,
 
 ESPEAK_API void espeak_SetSynthCallback(t_espeak_callback* SynthCallback);
 /* Must be called before any synthesis functions are called.
    This specifies a function in the calling program which is called when a buffer of
-   speech sound data has been produced. 
+   speech sound data has been produced.
 
 
    The callback function is of the form:
       start of the text.
 
    position_type:  Determines whether "position" is a number of characters, words, or sentences.
-      Values: 
+      Values:
 
    end_position:  If set, this gives a character position at which speaking will stop.  A value
       of zero indicates no end position.
          espeakCHARS_8BIT     The 8 bit ISO-8859 character set for the particular language.
          espeakCHARS_AUTO     8 bit or UTF8  (this is the default)
          espeakCHARS_WCHAR    Wide characters (wchar_t)
+         espeakCHARS_16BIT    16 bit characters.
 
       espeakSSML   Elements within < > are treated as SSML elements, or if not recognised are ignored.
 
       espeakENDPAUSE  If set then a sentence pause is added at the end of the text.  If not set then
          this pause is suppressed.
 
-   unique_identifier: message identifier; helpful for identifying later 
+   unique_identifier: message identifier; helpful for identifying later
      data supplied to the callback.
 
    user_data: pointer which will be passed to the callback function.
 
-   Return: EE_OK: operation achieved 
-           EE_BUFFER_FULL: the command can not be buffered; 
+   Return: EE_OK: operation achieved
+           EE_BUFFER_FULL: the command can not be buffered;
              you may try after a while to call the function again.
 	   EE_INTERNAL_ERROR.
 */
 
    For the other parameters, see espeak_Synth()
 
-   Return: EE_OK: operation achieved 
-           EE_BUFFER_FULL: the command can not be buffered; 
+   Return: EE_OK: operation achieved
+           EE_BUFFER_FULL: the command can not be buffered;
              you may try after a while to call the function again.
 	   EE_INTERNAL_ERROR.
 */
    If key_name is a single character, it speaks the name of the character.
    Otherwise, it speaks key_name as a text string.
 
-   Return: EE_OK: operation achieved 
-           EE_BUFFER_FULL: the command can not be buffered; 
+   Return: EE_OK: operation achieved
+           EE_BUFFER_FULL: the command can not be buffered;
              you may try after a while to call the function again.
 	   EE_INTERNAL_ERROR.
 */
 extern "C"
 #endif
 ESPEAK_API espeak_ERROR espeak_Char(wchar_t character);
-/* Speak the name of the given character 
+/* Speak the name of the given character
 
-   Return: EE_OK: operation achieved 
-           EE_BUFFER_FULL: the command can not be buffered; 
+   Return: EE_OK: operation achieved
+           EE_BUFFER_FULL: the command can not be buffered;
              you may try after a while to call the function again.
 	   EE_INTERNAL_ERROR.
 */
       espeakRANGE:   pitch range, range 0-100. 0-monotone, 50=normal
 
       espeakPUNCTUATION:  which punctuation characters to announce:
-         value in espeak_PUNCT_TYPE (none, all, some), 
+         value in espeak_PUNCT_TYPE (none, all, some),
          see espeak_GetParameter() to specify which characters are announced.
 
       espeakCAPITALS: announce capital letters by:
 
       espeakWORDGAP:  pause between words, units of 10mS (at the default speed)
 
-   Return: EE_OK: operation achieved 
-           EE_BUFFER_FULL: the command can not be buffered; 
+   Return: EE_OK: operation achieved
+           EE_BUFFER_FULL: the command can not be buffered;
              you may try after a while to call the function again.
 	   EE_INTERNAL_ERROR.
 */
 
    punctlist:  A list of character codes, terminated by a zero character.
 
-   Return: EE_OK: operation achieved 
-           EE_BUFFER_FULL: the command can not be buffered; 
+   Return: EE_OK: operation achieved
+           EE_BUFFER_FULL: the command can not be buffered;
              you may try after a while to call the function again.
 	   EE_INTERNAL_ERROR.
 */
    stream   output stream for the phoneme symbols (and trace).  If stream=NULL then it uses stdout.
 */
 
+#ifdef __cplusplus
+extern "C"
+#endif
+ESPEAK_API void espeak_TextToPhonemes(const void *text, char *buffer, int size, int textmode, int phonememode);
+/* Translates text into phonemes.  Call espeak_SetVoiceByName() first, to select a language.
+   text: The text to translate, terminated by a zero character.
+
+   buffer: Output buffer for the phoneme translation.
+
+   size: Size of the output buffer in bytes.
+
+   textmode: Type of character codes, one of:
+         espeakCHARS_UTF8     UTF8 encoding
+         espeakCHARS_8BIT     The 8 bit ISO-8859 character set for the particular language.
+         espeakCHARS_AUTO     8 bit or UTF8  (this is the default)
+         espeakCHARS_WCHAR    Wide characters (wchar_t)
+         espeakCHARS_16BIT    16 bit characters.
+
+   phonememode: bits0-3:
+      0= just phonemes.
+      1= include ties (U+361) for phoneme names of more than one letter.
+      2= include zero-width-joiner for phoneme names of more than one letter.
+      3= separate phonemes with underscore characters.
+
+	 bits 4-7:
+      0= eSpeak's ascii phoneme names.
+      1= International Phonetic Alphabet (as UTF-8 characters).
+*/
+
 #ifdef __cplusplus
 extern "C"
 #endif
 	unsigned char gender;  // 0=none 1=male, 2=female,
 	unsigned char age;     // 0=not specified, or age in years
 	unsigned char variant; // only used when passed as a parameter to espeak_SetVoiceByProperties
-	unsigned char xx1;     // for internal use 
+	unsigned char xx1;     // for internal use
 	int score;       // for internal use
 	void *spare;     // for internal use
 } espeak_VOICE;
 /* Searches for a voice with a matching "name" field.  Language is not considered.
    "name" is a UTF8 string.
 
-   Return: EE_OK: operation achieved 
-           EE_BUFFER_FULL: the command can not be buffered; 
+   Return: EE_OK: operation achieved
+           EE_BUFFER_FULL: the command can not be buffered;
              you may try after a while to call the function again.
 	   EE_INTERNAL_ERROR.
 */
    function returns, the audio output is fully stopped and the synthesizer is ready to
    synthesize a new message.
 
-   Return: EE_OK: operation achieved 
+   Return: EE_OK: operation achieved
 	   EE_INTERNAL_ERROR.
 */
 
 #endif
 ESPEAK_API espeak_ERROR espeak_Synchronize(void);
 /* This function returns when all data have been spoken.
-   Return: EE_OK: operation achieved 
+   Return: EE_OK: operation achieved
 	   EE_INTERNAL_ERROR.
 */
 
 #endif
 ESPEAK_API espeak_ERROR espeak_Terminate(void);
 /* last function to be called.
-   Return: EE_OK: operation achieved 
+   Return: EE_OK: operation achieved
 	   EE_INTERNAL_ERROR.
 */
 

src/synthdata.cpp

 #include "translate.h"
 #include "wave.h"
 
-const char *version_string = "1.47.07b  25.Apr.13";
+const char *version_string = "1.47.08  26.Apr.13";
 const int version_phdata  = 0x014701;
 
 int option_device_number = -1;

src/translate.h

 #define FLAG_ONLY           0x4000
 #define FLAG_ONLY_S         0x8000
 #define FLAG_STEM          0x10000  // must have a suffix
-#define FLAG_ATEND         0x20000  /* use this pronunciation if at end of clause */
+#define FLAG_ATEND         0x20000  // use this pronunciation if at end of clause
 #define FLAG_ATSTART       0x40000  // use this pronunciation if at start of clause
+#define FLAG_NATIVE        0x80000  // not if we've switched translators
 #define FLAG_LOOKUP_SYMBOL 0x40000000  // to indicate called from Lookup()
 
 #define BITNUM_FLAG_ALLCAPS   0x2a