[1.46.42]

Command line option -X now lists *_list flags by name.


git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@308 d46cf337-b52f-0410-862d-fd96e6ae7743

dictsource/dict_phonemes

 ts   v    x    z    
 
 
-Dictionary en_dict  2013-03-02
+Dictionary en_dict  2013-03-06
 
 0    0#   3    3:   @    @-   @2   @5   
 @L   a    a#   A:   A@   aa   aI   aI3  
 t.   th   th.  v    w    x    z    
 
 
-Dictionary ta_dict  2013-03-03
+Dictionary ta_dict  2013-03-05
 
 a    a:   aI   aU   e    E    e:   i    
 I    i:   o    o:   u    U    u2   u:   
 t    tS   v    w    x    
 
 
-Dictionary nl_dict  2013-03-03
+Dictionary nl_dict  2013-03-05
 
 8    @    @-   a    A    a:   A~   e    
 E    e#   E2   e:   EI   eU   i    I    
 y    Y:   yU   
 
 *    :    ;    b    d    D    dZ   f    
-g    h    j    k    l    m    n    N    
-n^   p    Q    r    s    S    t    tS   
-v    v#   w    x    z    Z    
+g    h    j    k    l    L    m    n    
+N    n^   p    Q    r    s    S    t    
+tS   v    v#   w    x    z    Z    
 
 
 Dictionary no_dict  2011-03-07
 ts;  v    w    x    z    Z    Z;   
 
 
-Dictionary pt_dict  2013-03-02
+Dictionary pt_dict  2013-03-06
 
 &    &/   &U~  &~   @    @-   a    A    
 aI   aU   e    E    eI   EI   eU   EU   
 w    x    z    Z    
 
 
-Dictionary ro_dict  2011-07-21
+Dictionary ro_dict  2013-03-04
 
 @    @-   @I   @U   a    aI   aU   e    
 ea   eI   eo   eU   i    i/   I^   iI   

dictsource/en_list

 // $alt1   use strong ending (-table, -inal)
 // $alt2   use weak ending
 // $alt3   use weak first vowel, a->[a#], e->[@], o->[0#]
+// $alt6   'to'
 
 
 // letter names, as they are spoken as a single character in text
 _bn	bEng'A:li
 _gur	g'U@mUki
 _gu	gu:dZ3r'A:ti
-_or	0r'i@
+_or	O:r'ia:
 _ta	t'amI2l
+_te	t'Elugu
 _kn	k'A:na#d@
 _ml	maleI'A:l@m
-_si	sI2nh'A:l@
+_si	s'INh@l@
 _th	t'aI
 _lo	l'aU
 _ti	tI2b'Et@n
 (for a while)    f@@||w'aIl
 (for one)        fO@w0n		$2 $atend
 
-to               t@5		$verbf    // @ change to U before vowel
+to               t@5		$verbf $alt6   // @ change to U before vowel
 (to be)          t@b%i		$pastf
 (to be)          t@||'bi:	$atend
 (to to)          %tU_t@5	$verbf
 thy              ,DaI
 thine            ,DaIn
 
+(i had)          aI||h'ad	$atend
+(he had)         hi:||h'ad	$atend
+
 me               ,mi:		$only
 me               mi:		$atstart $atend
 him              ,hIm		$only

dictsource/en_rules

             h (_
         _B) h
          @) hammed       h'amId
+         d) h (am_
          A) han (_       han          // places
          e) han (_       h@n
             ha  (st      heI

dictsource/ml_rules

 // A means vowel letters (not vowel signs)
 // B means a combining vowel sign or a virama
 
+// change (k t. t p) to (g d. d b) if followed by a vowel, unless at start of a word or preceeded by an unvoiced stop consonant ??
+.L01 _ ക്  ച്  ട്  ത്  പ്  
+
 
 .replace
 ൦	0	// Convert Malayalam numbers
 
 
 .group ക
-        ക          ka
-        ക (B       k
+        ക്          k
+   L01) ക          ka
+   L01) ക (B       k
+        ക          ga
+        ക (B       g
+        ക്ക          k:a
+        ക്ക (B       k:
 
 .group ഖ
         ഖ          kha
         ഞ (B       n^
 
 .group ട
-        ട          t.a
-        ട (B       t.
+        ട്          t.
+   L01) ട          t.a
+   L01) ട (B       t.
+        ട          d.a
+        ട (B       d.
+        ട്ട          t.:a
+        ട്ട (B       t.:
 
 .group ഠ
         ഠ          th.a
         ണ (B       n.
 
 .group ത
-        ത          ta
-        ത (B       t
+        ത്          t
+   L01) ത          ta
+   L01) ത (B       t
+        ത          da
+        ത (B       d
+        ത്ത          t:a
+        ത്ത (B       t:
 
 .group ഥ
         ഥ          tha
         ന (B       n
 
 .group പ
-        പ          pa
-        പ (B       p
+        പ്          p
+   L01) പ          pa
+   L01) പ (B       p
+        പ          ba
+        പ (B       b
+        പ്പ          p:a
+        പ്പ (B       p:
 
 .group ഫ
         ഫ          pha

dictsource/nl_list

 
 // This file in UTF8 encoded
 
+// $alt2, remove s from plural, remove en from plural
 // Use $alt3  where 'ng' and 'nk' is 'n+g' not [N]
 
 // Letters
 augustus	VUQ'8st8s
 bacterie	bAkt'I:ri
 banengids	$alt3
+bascule		$2
 beha            be:_h'a:
 beige		bEZ@
 benard		b@nArt
 december	$2
 dieet		$2
 diesel		diz@l
+discipline	$3
+//discussie	$2
 divers		div'Ers
 diftongering    dIft,ON'e:rIN
 doordat		$2
 kadaster	$2
 mangat		$alt3
 materie		ma:t'e:ri
+maatregel	ma:tre:Q@l $alt2
 maxima		$1
 meteen		mEt'e:n
 metro		me:tro:

dictsource/nl_rules

 // Suffix rule: SUFX_E  eg. S2e  means double the vowel before the final consonant
 
 .L01  g k
+.L02  _ t_ d_ de_ den_  // verb endings
 
 
 .group a
           &) bare (_S4    =ba:r@
           _) bij (C@P3    b'EI
              bij (zond    b%i
+          _) bijeen       bEI'e:n
           _) binnen (P6   b'In@n
           _) boven (P5    b'o:v@n
           _) brand        br'And
          ig) e (_+S1qd    @
 
              ea (_        'e:a:
-             eau          o: //words with French origin
+             eau          'o: //words with French origin
+//             eau (_       'o:
           @) ee (_        'e:
           @) ee (X_       'e:
           @) ee (r_       'I
           &) el (ingen_$p_alt =@l
 
           @) elen (_      'e:l@n
+          @) el (en_$p_alt2 @L
           &) elijks       @l@ks
           &) em (_        @m
           &) e (ment      @
           @) etten (_S3   @n 
           @) es    (_$p_alt @s  // for wors which end on [@] with multiple on 's'.
 
-             eau (_       'o:
              euille       @'y
              eum (_$w_alt 'e:8m
        @mus) eum (_       ,e:8m  // compounds
              ie           i
              ieuw         iw
              iee          ie:
+             ii           i
 
           d) ie (_        'i
           g) ie (_        'i
           m) ie (_        'i
           p) ie (_        'i
+         ss) ie (_        =i
              ieus (_      ij'Y:s
              ieuze (_     ij'Y:z@
              ieuzer (_    ij'Y:z@r
              io (nage     iO
              io (de_       i'o:
              iodes (_      i'o:d@s
+          _) in (scrip     In
              is (_        Is
           @) isch (_      =is
           @) ische (_     =is@
         @@C) o (_         =o:
              obsc (u      opsk
              oe (X_       'u
+             office       Of@s
              oir (_       'v#a:r
          pl) oi (t        v#A
           c) oi (ffu      v#A
           _) oost (@P4    o:st
           _) op (@@P2     'Op
           _) open (@P4    'o:p@n
+          _) op (enen     'o:p
           _) openings (P8 'o:p@nINs
           _) op (era      'o:p
           _) op (tion     Op
           A) s (A         z
              s (b         z
              s (d         z
-             sc           s
+             sc (e        s
+             sc (i        s
              sch          sx
              sch (_       s
              sch (e_      s
           a) s (_         s
           @) s (_$w_alt   =s
 
-         _) samen(@P5     s'a:m@n
+          _) samen(@P5    s'a:m@n
+          _) scan (L02    skEn
+             science      saI@ns
              sc (oop      sk
              sc (opi      sk
           &) sel (_       =s@l
           &) selen (_     =s@l@n
+             service      s8v@s
           @) si (o        S
           A) si (o        Z
           @) ssi (o       S

dictsource/pt_list

 abordo		$alt
 aborto		$alt  $verb
 acerto		$alt2 $noun
-acordo		$alt $verb
 acertos		$alt2
+acordo		$alt $verb
 acervo		$alt2
 adepto		$alt
 adereço		$alt2
 aterro		$alt2 $noun
 atmosfera	$alt
 atrozes		$alt
-austera		$alt
-austero		$alt
 autora		$alt2
 avesso	$alt2
 axé		aSE
 cerca	$alt
 cerca	$alt2 $noun
 cerco		$alt $verb
-cesta		$alt2
-cesto		$alt2
 chamego		$alt2
 chefe		$alt
 checam	$alt
 contivermos	$alt
 controle		$alt2 $noun
 controlo		$alt2 $noun
-copa		$alt
-copo		$alt
 coreia		$alt
 corneta		$alt2
 cornos		$alt
 coubermos	$alt
 cratera		$alt
 crede		$alt2
-cresce		$alt
-crescem		$alt
 der		$alt
 derdes	$alt
 desapego	$alt2 $noun
 endereço	$alt2
 endereço	$alt  $verb
 enforco		$alt
-enforque		$alt
-enforquem	$alt
 engordo		$alt
-ensopa		$alt
-ensopam	$alt
-ensope		$alt
-ensopem	$alt
-ensopo	$alt
 enredo		$alt  $verb
 enterro		$alt2 $noun
 enumerem	$alt
 ideia		$alt
 insonoro	$alt
 interesse		$alt2 $noun
-joga		$alt
-jogam$alt
-jogo	$alt $verb
-jogos	$alt
-jogue		$alt
-joguem	$alt
+jogo	$alt2 $noun
 joguete		$alt2
 lambreta	$alt2
 lebre		$alt
-líderes		$alt2
 lopes		$alt
 maior	$alt
 maleta		$alt2
 miolo		$alt2
 miolos	$alt
 moeda		$alt
-modo		$alt
 modos		$alt
 molho		$alt2 $noun
 monera		$alt
 relevo		$alt2 $noun
 remorso		$alt
 reprovo		$alt
-resolve		$alt
-resolvem	$alt
 retorno	$alt $verb
 rixa		x'iS&
 rocha		$alt
 rock		$alt
-rogas		$alt
 rola		$alt2 $noun
 rolo		$alt2 $noun
-rosna		$alt
-rosnam	$alt
-rosne		$alt
-rosnem	$alt
-rosno	$alt
 rumores		$alt2
 saleta		$alt2
 sarjeta		$alt
 ?1 senhora	$alt2
 senhores	$alt2
 sinopse		$alt
-sobe		$alt
-sobem	$alt
 soco		$alt2
 soco		$alt $verb
-sofrem		$alt
-sofre		$alt
-sofreste	$alt2
 sogra		$alt
-sopre		$alt
-soprem	$alt
-sopro		$alt $verb
+sopro		$alt2 $noun
 soro		$alt2
 sorvete		$alt2
 sossego     $alt2 $noun
 tivermos	$alt
 tocha		$alt
 toga		$alt
-tola		$alt2
-tolo		$alt2
-tolhe		$alt
-tolhem	$alt
 topo		$alt  $verb
 torno		$alt  $verb
 torre	$alt2 $noun

dictsource/pt_rules

  _estr) ei (L03_   EI
   _inv) e (jL04_   E
   _dec) e (pL04_   E
-    _s) e (quL02_  E
     _p) e (quL02_  E
+    _s) e (quL02_  E
      l) e (que_    E
   _sug) e (rL02_   E
   _imp) e (rL03_   E
     _f) e (rvL03_  e
     _p) e (sL01_   E
     _d) e (scL02_  E
+   _cr) e (scL02_  E
     _p) e (scL03_  E
     _m) e (sclL04  E
     _l) e (ssL02_  e
   _esp) e (ssL03_  e
+    _c) e (stL03_  e
     _m) e (tL03_   e
+   _pr) e (tL03_   e
   _com) e (tL03_   e
   _rem) e (tL03_   e
   prom) e (tL03_   e         // com-
  _derr) e (tL03_   e
 _obsol) e (tL03_   e
-   _pr) e (tL03_   e
   _sol) e (trL03_  E
   _pen) e (trL04_  E
-   scr) e (vL03_   e         // e- in- pre- ree- de-
     _d) e (vL03_   e
+   scr) e (vL03_   e         // e- in- pre- ree- de-
   _atr) e (vL03_   e
-_embel) e (zL04_   E
     pr) e (zL04_   E         / des- menos-
+_embel) e (zL04_   E
 _acont) e (çL01_   e
 _arref) e (çL01_   e
    par) e (çL03_   e         // a- re- com- rea- desa-
   _mer) e (çL03_   e
   _pad) e (çL03_   e         // com-
   _per) e (çL03_   e
-_rejuven) e (çL03_ e
   conh) e (çL03_   e         // re-
  _esqu) e (çL03_   e
  _forn) e (çL03_   e
 _esclar) e (çL03_  e
 _estrem) e (çL03_  e
 _fortal) e (çL03_  e
+_rejuven) e (çL03_ e
 //endsort
 
 
 
 //sort
  _engl) o (bL01_   O
+    _s) o (bL02_   O
   _esn) o (bL04_   O
   _deb) o (chL04_  O
 _desabr) o (chL04_ O
+    _s) o (frL02_  O
+    _j) o (gL04_   O
     _r) o (gL04_   O
   _adv) o (gL04_   O
   _ref) o (gL04_   O
    _ap) oi (L01_   OI
-  _esc) o (lhL02_  O
+    _t) o (lL03_   o
+     c) o (lhL02_  O         / es- en- re-
+    _t) o (lhL02_  O
     _m) o (lhL04_  O
+   env) o (lvL02_  O         / des-
   _abs) o (lvL02_  O
-  _env) o (lvL02_  O
+  _res) o (lvL02_  O
   _rev) o (lvL02_  O
-_desenv) o (lvL02_ O
   _ent) o (pL02_   O
+    _c) o (pL03_   O
+  _ens) o (pL04_   O
+    _s) o (prL04_  O
+  _enf) o (rquL02_ O
      c) o (rrL01_  o         // escorra, incorra, corra, recorra
     _m) o (rrL01_  o
+    _r) o (snL04_  O
     _c) o (spL02_  O
      p) ostos (_   Ost=Us#   // dispostos etc.
-    _s) o (bL02_   O
-    _s) o (prL02_  O
 //endsort
 
 .group ô

dictsource/ro_list

 vinovăţi        $alt
 vinovăţii       $4
 virgulă		$1
-vis-a-vis vizav'i
+(vis-a vis) vizav'i
 viscol $1
 vitreg $1
 vitrege $1

dictsource/ro_rules

    ţi-) i (_       _         // combined with previous word as [iI]
     Cr) i (_+++    i
     V&) i (N_      'i        // verb
+    
+     @) i (A$p_alt1 i        // i is not [j]
     ăC) ia (_      'ia
     âC) ia (_      'ia
     ăC) ie (_      'ie

dictsource/ta_list

 ½	araI
 ¾	mukka:l
 
+// alphabet names
+_ar	arVbU
+_cyr	sirillik
+_he	;ebire:jVm
+
+
 // abbreviations
 ரூ	ru:ba:j  $dot $only
 (பி . ஏ)	bije: $dot

phsource/compile_report

                 [l/] fr
 l/l_@           [l/3] base
                 [l/] fr
-l/l@            [l#] base
-                [X)]] base
+l/l@            [¨\-] base
+                [l#] base
                 [l] fr
                 [l/2] fr
                 [K] nso
 l/L2_uL         [l/2] base
 l/l_3           [l/] de
 l/l_4           [ll] sq
-l/la            [l#] base
-                [X)]] base
+l/la            [¨\-] base
+                [l#] base
                 [l] fr
                 [l/2] fr
                 [K] nso
                 [K] tn
 l/l_a           [l/3] base
                 [l/] fr
-l/le            [l#] base
-                [X)]] base
+l/le            [¨\-] base
+                [l#] base
                 [l] fr
                 [l/2] fr
                 [K] nso
                 [&:] af
 l/l_front       [L] sq
 l/l_front_      [l/4] sq
-l/li            [l#] base
-                [X)]] base
+l/li            [¨\-] base
+                [l#] base
                 [l] fr
                 [l/2] fr
                 [l] zh
 ll/_ll          [L] base
 l/l_long        [l] base
                 [l] fr
-l/lo            [l#] base
-                [X)]] base
+l/lo            [¨\-] base
+                [l#] base
                 [l/2] fr
                 [K] nso
                 [K] tn
                 [l;] lt
                 [l] ru
                 [l^] ru
-l/lu            [l#] base
-                [X)]] base
+l/lu            [¨\-] base
+                [l#] base
                 [l] fr
                 [l/2] fr
                 [K] nso

platforms/windows/espeakedit/src/speech.h

 #define mkdir(p1,p2) mkdir(p1)
 
 #define PATHSEP  '\\'
-#define USE_PORTAUDIO
+//#define USE_PORTAUDIO
 //#define USE_NANOSLEEP
 #define NO_VARIADIC_MACROS
 #define __cdecl 

platforms/windows/espeakedit/src_copy/speech.h

 #define mkdir(p1,p2) mkdir(p1)
 
 #define PATHSEP  '\\'
-#define USE_PORTAUDIO
+//#define USE_PORTAUDIO
 //#define USE_NANOSLEEP
 #define NO_VARIADIC_MACROS
 #define __cdecl 

platforms/windows/make_espeak.iss

 
 [Setup]
 AppName=eSpeak
-AppVerName=eSpeak version 1.46.31
+AppVerName=eSpeak version 1.46.42
 AppCopyright=Licensed under GNU General Public License version 3.   (See file License.txt for details).
 WindowVisible=yes
 
 
 [InstallDelete]
 Type: files; Name: "{app}\espeak.dll"
+Type: files; Name: "{app}\espeak-data\voices\*"
+Type: filesandordirs; Name: "{app}\espeak-data\voices\en"
+Type: filesandordirs; Name: "{app}\espeak-data\voices\asia"
+Type: filesandordirs; Name: "{app}\espeak-data\voices\europe"
+Type: filesandordirs; Name: "{app}\espeak-data\voices\other"
 Type: filesandordirs; Name: "{app}\espeak-data\voices\test"
 Type: filesandordirs; Name: "{app}\docs"
 
   $41: Result := 'sw';
   //$44: Result := 'tt';
   $49: Result := 'ta';
+  $46: Result := 'pa';
   $4b: Result := 'kn';
   //$50: Result := 'mn';
   $52: Result := 'cy';
   'an': value := $40a;   // Aragon, use code for Spanish
   'az': value := $42c;
   'bg': value := $402;
+  'bn': value := $445;
   'bs': value := $41a;   // should be $141a but Jaws crashes on startup
   'ca': value := $403;
   'cs': value := $405;
   'ne': value := $461;
   'nl': value := $413;
   'no': value := $414;
+  'pa': value := $446;
   'pl': value := $415;
   'pt': value := $416;
   'ro': value := $418;

src/compiledict.cpp

 }   /* end of LookupMnemValue */
 
 
-char *print_dictionary_flags(unsigned int *flags)
-{//==============================================
-	static char buf[20];
+void print_dictionary_flags(unsigned int *flags, char *buf, int buf_len)
+{//========================================================================
+	int stress;
+	int ix;
+	const char *name;
+	int len;
+	int total = 0;
+
+    buf[0] = 0;
+	if((stress = flags[0] & 0xf) != 0)
+	{
+	    sprintf(buf, "%s", LookupMnemName(mnem_flags, stress + 0x40));
+	    total = strlen(buf);
+	    buf += total;
+	}
 
-	sprintf(buf,"%s  0x%x/%x",LookupMnemName(mnem_flags,(flags[0] & 0xf)+0x40), flags[0], flags[1]);
-	return(buf);
+    for(ix=8; ix<64; ix++)
+    {
+        if(((ix < 30) && (flags[0] & (1 << ix))) || ((ix >= 0x20) && (flags[1] & (1 << (ix-0x20)))))
+        {
+            name = LookupMnemName(mnem_flags, ix);
+            len = strlen(name) + 1;
+            total += len;
+            if(total >= buf_len)
+                continue;
+            sprintf(buf, " %s", name);
+            buf += len;
+        }
+    }
 }
 
 

src/dictionary.cpp

 int dictionary_skipwords;
 char dictionary_name[40];
 
-extern char *print_dictionary_flags(unsigned int *flags);
+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
 		{
 			w = word;
 		}
-		while(*p == *w)
+		while((*p == *w) && (*w != 0))
 		{
 			w++;
 			p++;
 							p += (wc_bytes-1);
 						}
 
-						if(((alphabet = AlphabetFromChar(letter)) != NULL) && (alphabet->language != 0) && (alphabet->offset != tr->letter_bits_offset))
+						if(((alphabet = AlphabetFromChar(letter)) != NULL) && (alphabet->flags & AL_WORDS) && (alphabet->offset != tr->letter_bits_offset))
 						{
 						    // switch to the nominated language for this alphabet
                             sprintf(phonemes,"%c%s",phonSWITCH, WordToString2(alphabet->language));
 		}
 	}
 
-	// any language specific changes ?
-	ApplySpecialAttribute(tr,phonemes,dict_flags0);
 	memcpy(p_start,word_copy,strlen(word_copy));
 
 	return(0);
 }  // end of ApplySpecialAttribute2
 
 
-void ApplySpecialAttribute(Translator *tr, char *phonemes, int dict_flags)
-{//=======================================================================
-// Amend the translated phonemes according to an attribute which is specific for the language.
-	int len;
-	char *p_end;
-
-	if((dict_flags & (FLAG_ALT_TRANS | FLAG_ALT2_TRANS)) == 0)
-		return;
-
-	len = strlen(phonemes);
-	p_end = &phonemes[len-1];
-
-	switch(tr->translator_name)
-	{
-#ifdef deleted
-// this is now done in de_rules
-	case L('d','e'):
-		if(p_end[0] == PhonemeCode2('i',':'))
-		{
-			// words ends in ['i:], change to [=I@]
-			p_end[-1] = phonSTRESS_PREV;
-			p_end[0] = PhonemeCode('I');
-			p_end[1] = phonSCHWA;
-			p_end[2] = 0;
-		}
-		break;
-#endif
-
-	case L('r','o'):
-		if(p_end[0] == PhonemeCode('j'))
-		{
-			// word end in [j], change to ['i]
-			p_end[0] = phonSTRESS_P;
-			p_end[1] = PhonemeCode('i');
-			p_end[2] = 0;
-		}
-		break;
-	}
-}  // end of ApplySpecialAttribute
-
 
 
 //=============================================================================================
 	const char *word1;
 	int wflags = 0;
 	char word_buf[N_WORD_BYTES+1];
+	char dict_flags_buf[80];
 
 	if(wtab != NULL)
 	{
 			if(tr->expect_verb || (tr->expect_verb_s && (end_flags & FLAG_SUFX_S)))
 			{
 				// OK, we are expecting a verb
+				if((tr->translator_name == L('e','n')) && (tr->prev_dict_flags[0] & FLAG_ALT6_TRANS) && (end_flags & FLAG_SUFX_S))
+				{
+				    // lang=en, don't use verb form after 'to' if the word has 's' suffix
+				    continue;
+				}
 			}
 			else
 			{
 		if(dictionary_flags & FLAG_ALT2_TRANS)
 		{
 			// language specific
-			if((tr->translator_name == L('h','u')) && !(tr->prev_dict_flags & FLAG_ALT_TRANS))
+			if((tr->translator_name == L('h','u')) && !(tr->prev_dict_flags[0] & FLAG_ALT_TRANS))
 				continue;
 		}
 
 		{
 			if(option_phonemes == 2)
 			{
-				fprintf(f_trans,"Flags:  %s  %s\n",word1,print_dictionary_flags(flags));
+			    print_dictionary_flags(flags, dict_flags_buf, sizeof(dict_flags_buf));
+				fprintf(f_trans,"Flags:  %s  %s\n", word1, dict_flags_buf);
 			}
 			return(0);    // no phoneme translation found here, only flags. So use rules
 		}
 			if(textmode == translator->langopts.textmode)
 			{
 				// only show this line if the word translates to phonemes, not replacement text
-				if((dictionary_skipwords) && (wtab != NULL))
+				if((dictionary_flags & FLAG_SKIPWORDS) && (wtab != NULL))
 				{
 					// matched more than one word
 					// (check for wtab prevents showing RULE_SPELLING byte when speaking individual letters)
 					memcpy(word_buf,word2,word_end-word2);
 					word_buf[word_end-word2-1] = 0;
-					fprintf(f_trans,"Found: '%s %s",word1,word_buf);
+					fprintf(f_trans,"Found: '%s %s\n",word1,word_buf);
 				}
 				else
 				{
 					fprintf(f_trans,"Found: '%s",word1);
 				}
-				fprintf(f_trans,"' [%s]  %s\n",ph_decoded,print_dictionary_flags(flags));
+			    print_dictionary_flags(flags, dict_flags_buf, sizeof(dict_flags_buf));
+				fprintf(f_trans,"' [%s]  %s\n", ph_decoded,dict_flags_buf);
 			}
 		}
 

src/espeak_command.cpp

     {
       goto text_error;
     }
- 
-  a_text = malloc( size );
+
+  a_text = malloc( size+1 );
   if (!a_text)
     {
       goto text_error;
     }
   memcpy(a_text, text, size);
-  
+
   a_command->type = ET_TEXT;
   a_command->state = CS_UNDEFINED;
   data = &(a_command->u.my_text);
     {
       goto msg_error;
     }
-   
+
   a_command->type = ET_TERMINATED_MSG;
   a_command->state = CS_UNDEFINED;
   data = &(a_command->u.my_terminated_msg);
     {
       goto char_error;
     }
- 
+
   a_command->type = ET_CHAR;
   a_command->state = CS_UNDEFINED;
   a_command->u.my_char.user_data = user_data;
     {
       goto param_error;
     }
- 
+
   a_command->type = ET_PARAMETER;
   a_command->state = CS_UNDEFINED;
   data = &(a_command->u.my_param);
-  data->parameter = parameter; 
+  data->parameter = parameter;
   data->value = value;
   data->relative = relative;
   a_error=0;
     {
       goto list_error;
     }
- 
+
   a_command->type = ET_PUNCTUATION_LIST;
   a_command->state = CS_UNDEFINED;
 
     {
       goto name_error;
     }
- 
+
   a_command->type = ET_VOICE_NAME;
   a_command->state = CS_UNDEFINED;
   a_command->u.my_voice_name = strdup( name);
     {
       goto spec_error;
     }
- 
+
   a_command->type = ET_VOICE_SPEC;
   a_command->state = CS_UNDEFINED;
   {
 	  break;
 
 	case ET_TERMINATED_MSG:
-	  { 
+	  {
 	    // if the terminated msg is pending,
-	    // it must be processed here for informing the calling program 
+	    // it must be processed here for informing the calling program
 	    // that its message is finished.
-	    // This can be important for cleaning the related user data.	    
+	    // This can be important for cleaning the related user data.
 	    t_espeak_terminated_msg* data = &(the_command->u.my_terminated_msg);
 	    if (the_command->state == CS_PENDING)
 	      {
 	    free((void*)(the_command->u.my_voice_name));
 	  }
 	  break;
-	  
+
 	case ET_VOICE_SPEC:
 	  {
 		espeak_VOICE* data = &(the_command->u.my_voice_spec);
     case ET_TEXT:
       {
 	t_espeak_text* data = &(the_command->u.my_text);
-	sync_espeak_Synth( data->unique_identifier, data->text, data->size, 
-			   data->position, data->position_type, 
-			   data->end_position, data->flags, data->user_data);	
+	sync_espeak_Synth( data->unique_identifier, data->text, data->size,
+			   data->position, data->position_type,
+			   data->end_position, data->flags, data->user_data);
       }
       break;
 
     case ET_MARK:
       {
 	t_espeak_mark* data = &(the_command->u.my_mark);
-	sync_espeak_Synth_Mark( data->unique_identifier, data->text, data->size, 
-				data->index_mark, data->end_position, data->flags, 
+	sync_espeak_Synth_Mark( data->unique_identifier, data->text, data->size,
+				data->index_mark, data->end_position, data->flags,
 				data->user_data);
       }
       break;
 
     case ET_TERMINATED_MSG:
-      { 
+      {
 	t_espeak_terminated_msg* data = &(the_command->u.my_terminated_msg);
 	sync_espeak_terminated_msg( data->unique_identifier, data->user_data);
       }
       {
 	t_espeak_terminated_msg* data = &(the_command->u.my_terminated_msg);
 
-	SHOW("display_espeak_command > (0x%x) TERMINATED_MSG uid=%d, user_data=0x%x, state=%d\n", 
-	     the_command, data->unique_identifier, data->user_data, 
+	SHOW("display_espeak_command > (0x%x) TERMINATED_MSG uid=%d, user_data=0x%x, state=%d\n",
+	     the_command, data->unique_identifier, data->user_data,
 	     the_command->state);
       }
       break;
     case ET_PARAMETER:
       {
 	t_espeak_parameter* data = &(the_command->u.my_param);
-	SHOW("display_espeak_command > (0x%x) PARAMETER=%d, value=%d, relative=%d\n", 
+	SHOW("display_espeak_command > (0x%x) PARAMETER=%d, value=%d, relative=%d\n",
 	     the_command, data->parameter, data->value, data->relative);
       }
       break;

src/numbers.cpp

 		return(0);
 	}
 
-#ifdef deleted
-	if((ph_buf[0] == 0) && (tr->translator_name != L('e','n')))
-	{
-		// speak as English, check whether there is a translation for this character
-		SetTranslator2("en");
-		save_option_phonemes = option_phonemes;
-		option_phonemes = 0;
-		LookupLetter(translator2, letter, word[n_bytes], ph_buf, control & 1);
-		SelectPhonemeTable(voice->phoneme_tab_ix);  // revert to original phoneme table
-		option_phonemes = save_option_phonemes;
-
-		if(ph_buf[0] != 0)
-		{
-			sprintf(phonemes,"%cen",phonSWITCH);
-			return(0);
-		}
-	}
-#endif
-
     alphabet = AlphabetFromChar(letter);
     if(alphabet != current_alphabet)
     {
 						nextflags = TranslateWord(tr, &word_end[2], 0, NULL, NULL);
 					}
 
-if((tr->prev_dict_flags & FLAG_ALT_TRANS) && ((c2 == 0) || (wtab[0].flags & FLAG_COMMA_AFTER) || iswdigit(c2)))
+if((tr->prev_dict_flags[0] & FLAG_ALT_TRANS) && ((c2 == 0) || (wtab[0].flags & FLAG_COMMA_AFTER) || iswdigit(c2)))
 	ordinal = 0;   // TEST  09.02.10
 
 					if(nextflags & FLAG_ALT_TRANS)
 						if(word[-2] == '-')
 							ordinal = 0;   // eg. december 2-5. között
 
-						if(tr->prev_dict_flags & (FLAG_ALT_TRANS | FLAG_ALT3_TRANS))
+						if(tr->prev_dict_flags[0] & (FLAG_ALT_TRANS | FLAG_ALT3_TRANS))
 							ordinal = 0x22;
 					}
 				}
 		}
 	}
 
-	tr->prev_dict_flags = 0;
+	tr->prev_dict_flags[0] = 0;
+	tr->prev_dict_flags[1] = 0;
 	TranslateNumber(tr, &number_chars[2], p, flags, wtab, num_control);
 
 	if(tr->langopts.numbers & NUM_ROMAN_AFTER)

src/speech.h

 #endif
 
 #ifdef _ESPEAKEDIT
-#define USE_PORTAUDIO
+//#define USE_PORTAUDIO
 #define USE_ASYNC
 #define LOG_FRAMES      // write keyframe info to log-espeakedit
 #endif

src/synthdata.cpp

 #include "translate.h"
 #include "wave.h"
 
-const char *version_string = "1.46.41  03.Mar.13";
+const char *version_string = "1.46.42  06.Mar.13";
 const int version_phdata  = 0x014640;
 
 int option_device_number = -1;

src/tr_languages.cpp

 ALPHABET alphabets [] = {
     {"_el",    OFFSET_GREEK,    0x380, 0x3ff,  L('e','l'), AL_DONT_NAME | AL_NOT_LETTERS},
     {"_cyr",   OFFSET_CYRILLIC, 0x400, 0x52f,  0, 0},
-    {"_hy",    OFFSET_ARMENIAN, 0x530, 0x58f,  L('h','y'), 0},
+    {"_hy",    OFFSET_ARMENIAN, 0x530, 0x58f,  L('h','y'), AL_WORDS},
     {"_he",    OFFSET_HEBREW,   0x590, 0x5ff,  0, 0},
     {"_ar",    OFFSET_ARABIC,   0x600, 0x6ff,  0, 0},
     {"_dv",    OFFSET_THAANA,   0x780, 0x7bf,  0, 0},
-    {"_hi",    OFFSET_DEVANAGARI, 0x900, 0x97f,L('h','i'), 0},
+    {"_hi",    OFFSET_DEVANAGARI, 0x900, 0x97f,L('h','i'), AL_WORDS},
     {"_bn",    OFFSET_BENGALI,  0x0980, 0x9ff, 0, 0},
     {"_gur",   OFFSET_GURMUKHI, 0xa00, 0xa7f,  0, 0},
     {"_gu",    OFFSET_GUJARATI, 0xa80, 0xaff,  0, 0},
     {"_or",    OFFSET_ORIYA,    0xb00, 0xb7f,  0, 0},
-    {"_ta",    OFFSET_TAMIL,    0xb80, 0xbff,  L('t','a'), 0},
-    {"_te",    OFFSET_TELUGU,   0xc00, 0xc7f,  0, 0},
-    {"_kn",    OFFSET_KANNADA,  0xc80, 0xcff,  L('k','n'), 0},
-    {"_ml",    OFFSET_MALAYALAM,0xd00, 0xd7f,  L('m','l'), 0},
+    {"_ta",    OFFSET_TAMIL,    0xb80, 0xbff,  L('t','a'), AL_WORDS},
+    {"_te",    OFFSET_TELUGU,   0xc00, 0xc7f,  L('t','e'), 0},
+    {"_kn",    OFFSET_KANNADA,  0xc80, 0xcff,  L('k','n'), AL_WORDS},
+    {"_ml",    OFFSET_MALAYALAM,0xd00, 0xd7f,  L('m','l'), AL_WORDS},
     {"_si",    OFFSET_SINHALA,  0xd80, 0xdff,  0, 0},
     {"_th",    OFFSET_THAI,     0xe00, 0xe7f,  0, 0},
     {"_lo",    OFFSET_LAO,      0xe80, 0xeff,  0, 0},
     {"_ti",    OFFSET_TIBET,    0xf00, 0xfff,  0, 0},
     {"_my",    OFFSET_MYANMAR,  0x1000,0x109f, 0, 0},
-    {"_ka",    OFFSET_GEORGIAN, 0x10a0,0x10ff, L('k','a'), 0},
+    {"_ka",    OFFSET_GEORGIAN, 0x10a0,0x10ff, L('k','a'), AL_WORDS},
     {"_ko",    OFFSET_KOREAN,   0x1100,0x11ff, 0, 0},
     {"_eth",   OFFSET_ETHIOPIC, 0x1200,0x139f, 0, 0},
     {NULL, 0, 0, 0, 0, 0}

src/translate.cpp

                         {
                             // don't retranslate, use the original lookup result
                             strcpy(phonemes,phonemes2);
-
-                            // language specific changes
-                            ApplySpecialAttribute(tr,phonemes,dictionary_flags[0]);
                         }
                         else
                         {
 			pre_pause--;
 		}
 		tr->end_stressed_vowel = 0;   // forget about the previous word
-		tr->prev_dict_flags = 0;
+		tr->prev_dict_flags[0] = 0;
+		tr->prev_dict_flags[1] = 0;
 	}
 
 	if((option_capitals==1) && (word_flags & FLAG_FIRST_UPPER))
 			{
 				if(first_phoneme && tr->langopts.param[LOPT_IT_DOUBLING])
 				{
-					if(((tr->prev_dict_flags & FLAG_DOUBLING) && (tr->langopts.param[LOPT_IT_DOUBLING] & 1)) ||
+					if(((tr->prev_dict_flags[0] & FLAG_DOUBLING) && (tr->langopts.param[LOPT_IT_DOUBLING] & 1)) ||
 						(tr->end_stressed_vowel && (tr->langopts.param[LOPT_IT_DOUBLING] & 2)))
 					{
 						// italian, double the initial consonant if the previous word ends with a
 		ph_list2[max_stress_ix].synthflags |= SFLAG_PROMOTE_STRESS;
 	}
 
-	tr->prev_dict_flags = flags;
+	tr->prev_dict_flags[0] = flags;
 	return(flags);
 }  //  end of TranslateWord2
 
 	tr->expect_verb_s=0;
 	tr->phonemes_repeat_count = 0;
 	tr->end_stressed_vowel=0;
-	tr->prev_dict_flags = 0;
+	tr->prev_dict_flags[0] = 0;
+	tr->prev_dict_flags[1] = 0;
 
 	word_count = 0;
 	single_quoted = 0;

src/translate.h

 #define FLAG_ALT_TRANS      0x8000  // language specific
 #define FLAG_ALT2_TRANS    0x10000  // language specific
 #define FLAG_ALT3_TRANS    0x20000  // language specific
-#define FLAG_COMBINE       0x40000  // combine with the next word
+#define FLAG_ALT4_TRANS    0x40000  // language specific
+#define FLAG_ALT5_TRANS    0x80000  // language specific
+#define FLAG_ALT6_TRANS   0x100000  // language specific
 
+#define FLAG_COMBINE      0x800000  // combine with the next word
 #define FLAG_ALLOW_DOT  0x01000000  // ignore '.' after word (abbreviation)
 #define FLAG_NEEDS_DOT  0x02000000  // only if the word is followed by a dot
 #define FLAG_WAS_UNPRONOUNCABLE  0x04000000  // the unpronounceable routine was used
 #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_ATSTART       0x30000  // use this pronunciation if at start of clause
+#define FLAG_ATSTART       0x40000  // use this pronunciation if at start of clause
 
 #define BITNUM_FLAG_ALLCAPS   0x2a
 #define BITNUM_FLAG_HYPHENATED  0x2c
 // alphabet flags
 #define AL_DONT_NAME  0x01    // don't speak the alphabet name
 #define AL_NOT_LETTERS  0x02  // don't use the language for speaking letters
-#define AL_NOT_WORDS  0x04    // don't use the language to speak words
+#define AL_WORDS  0x04        // use the language to speak words
 
 
 #define N_LOPTS      21
 	unsigned char groups2_start[256];    // index into groups2
 	const short *frequent_pairs;   // list of frequent pairs of letters, for use in compressed *_list
 
-
 	int expect_verb;
 	int expect_past;    // expect past tense
 	int expect_verb_s;
 
 	int prepause_timeout;
 	int end_stressed_vowel;  // word ends with stressed vowel
-	int prev_dict_flags;     // dictionary flags from previous word
+	int prev_dict_flags[2];     // dictionary flags from previous word
 	int clause_terminator;
 } Translator;
 
 
 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);
-void ApplySpecialAttribute(Translator *tr, char *phonemes, int dict_flags);
 void ApplySpecialAttribute2(Translator *tr, char *phonemes, int dict_flags);
 void AppendPhonemes(Translator *tr, char *string, int size, const char *ph);
 

src/voices.cpp

 	const char *language_type;
 	char buf[200];
 	char path_voices[sizeof(path_home)+12];
-	char langname[4];
 
 	int stress_amps[8];
 	int stress_lengths[8];
 	int stress_add[8];
 	char names[8][40];
+	char name2[80];
+	const char *voice_dir;
 
 	int pitch1;
 	int pitch2;
 	static char voice_name[40];        // voice name for current_voice_selected
 	static char voice_languages[100];  // list of languages and priorities for current_voice_selected
 
+    // which directory to look for a named voice
+    static const char *voices_asia =
+        "fa fa-pin hi hy hy-west id ka kn ku ml ne pa ta tr vi vi-hue zh zh-yue ";
+    static const char *voices_europe =
+        "bg bs ca cs cy da el es et fi fr-be hr hu is it lt lv mk nl no pl pt-pt ro ru sk sq sr sv ";
+
 	strcpy(voicename,vname);
 	if(voicename[0]==0)
 		strcpy(voicename,"default");
 	else
 	{
 		sprintf(path_voices,"%s%cvoices%c",path_home,PATHSEP,PATHSEP);
-		sprintf(buf,"%s%s",path_voices,voicename);
+		sprintf(buf,"%s%s",path_voices,voicename);  // first, look in the main voices directory
 
 		if(GetFileLength(buf) <= 0)
 		{
-			// look for the voice in a sub-directory of the language name
-			langname[0] = voicename[0];
-			langname[1] = voicename[1];
-			langname[2] = 0;
-			sprintf(buf,"%s%s%c%s",path_voices,langname,PATHSEP,voicename);
-
-			if(GetFileLength(buf) <= 0)
-			{
-				// look in "extra" sub-directory
-				sprintf(buf,"%sextra%c%s",path_voices,PATHSEP,voicename);
-
-				if(GetFileLength(buf) <= 0)
-				{
-					// look in "test" sub-directory
-					sprintf(buf,"%stest%c%s",path_voices,PATHSEP,voicename);
-				}
-			}
+		    // then look in the appropriate subdirectory
+		    if((voicename[0]=='m') && (voicename[1]=='b'))
+		    {
+		        voice_dir = "mb";   // mbrola voices
+		    }
+		    else
+		    {
+                sprintf(name2, "%s ", voicename);
+                if(strstr(voices_europe, voicename) != NULL)
+                    voice_dir = "europe";
+                else
+                if(strstr(voices_asia, voicename) != NULL)
+                    voice_dir = "asia";
+                else
+                    voice_dir = "other";
+
+                sprintf(buf,"%s%s%c%s", path_voices,voice_dir,PATHSEP,voicename);
+
+                if(GetFileLength(buf) <= 0)
+                {
+                    // if not found, look in "test" sub-directory
+                    sprintf(buf,"%stest%c%s",path_voices,PATHSEP,voicename);
+                }
+		    }
 		}
 	}
 

src/wave.cpp

 // notdef USE_PORTAUDIO
 
 
-int wave_init(int srate) {}
+int wave_init(int srate) {return 1;}
 void* wave_open(const char* the_api) {return (void *)1;}
 size_t wave_write(void* theHandler, char* theMono16BitsWaveBuffer, size_t theSize) {return theSize;}
 int wave_close(void* theHandler) {return 0;}

src/wave_pulse.cpp

 // notdef USE_PULSEAUDIO
 
 
-int wave_init() {}
+int wave_init(return 1;) {}
 void* wave_open(const char* the_api) {return (void *)1;}
 size_t wave_write(void* theHandler, char* theMono16BitsWaveBuffer, size_t theSize) {return theSize;}
 int wave_close(void* theHandler) {return 0;}

src/wave_sada.cpp

 #else
 // notdef USE_SADA
 
-init wave_init() {}
+init wave_init() {return 1;}
 void* wave_open(const char* the_api) {return (void *)1;}
 size_t wave_write(void* theHandler, char* theMono16BitsWaveBuffer, size_t theSize) {return theSize;}
 int wave_close(void* theHandler) {return 0;}