[1.31.17]

*_rules: New option, (Pb to specify a character such as apostrophe which splits a word into two parts (used for lang-tr).
 

git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@150 d46cf337-b52f-0410-862d-fd96e6ae7743

dictsource/dict_phonemes

@@ -88,13 +88,13 @@ r    R    s    t    ts   v    z    Z
 
 Dictionary es_dict
 
-@-   a    aI   e    E    eI   eU   i    
-o    O    oI   u    
+@-   a    aI   aU   e    E    eI   eU   
+i    o    O    oI   u    
 
 *    **   :    b    B    d    D    f    
-g    j    J    J^   k    l    l^   m    
-n    n^   p    Q    r    R    s    t    
-T    tS   v    v#   w    x    z    
+g    j    J    J^   k    l    m    n    
+n^   p    Q    r    R    s    t    T    
+tS   v    v#   w    x    z    
 
 
 Dictionary fi_dict
@@ -397,10 +397,9 @@ Dictionary tr_dict
 a    e    E    i    I    o    O    u    
 u#   W    y    Y    
 
-:    ;    b    d    dZ   f    g    h    
-j    k    l    m    n    p    Q    r    
-R    s    S    t    tS   v    w    z    
-Z    
+b    c    d    dZ   f    g    h    j    
+J    k    l    m    n    p    r    R    
+s    S    t    tS   v    z    Z    
 
 
 Dictionary ku_dict

dictsource/es_list

@@ -10,63 +10,63 @@ _#32	Esp'aTjo
 
 // names of symbols
 _.	punto
-_,	kOma
-_;	p,untOik'Oma
+_,	koma
+_;	p,untoik'oma
 _:	d,Osp'untOs
 _!	TE*R'aRaDmi**aTj'On
-_?	TE*R'aRintE*R,OQaTj'On
+_?	TE*R'aRintE*R,oQaTj'On
 _¡	aB@-*'iRaDmi**aTj'On
-_¿	aB@-*'iRintE*R,OQaTj'On
+_¿	aB@-*'iRintE*R,oQaTj'On
 _=	iQw'al
-_<	mEn'ORke
+_<	men'ORke
 _>	maJ^'ORke
-_'	apOst@-*'Ofo
-_"	kOm'iJ^as
+_'	apOst@-*'ofo
+_"	kom'iJ^as
 _+	s'iQnOdesum'aR
-_$	dOlar
-_#	almOaD'il^a
-_*	astE*'isko
+_$	dolar
+_#	almoaD'iJ^a
+_*	aste*'isko
 _-	gJ^On
 __	suB@-*aJ^'aDo
 _/	ba*Ra
 _\	b'a*RaimbERt'iDa
 _`	aT'EntOg@-*'aBe
-_(	,aB@-*epa**'EntEsis
-_)	Tj,E*Rapa**'EntEsis
-_[	,aB@-*ekORtS'Ete
-_]	Tj,E*RakORtS'Ete
+_(	,aB@-*epa**'Entesis
+_)	Tj,E*Rapa**'Entesis
+_[	,aB@-*ekORtS'ete
+_]	Tj,E*RakORtS'ete
 _{	,aB@-*eJ^'aBe
 _}	Tj,E*RaJ^'aBe
-_«	kOm'iJ^as||iTkJ^'ERDas
-_»	kOm'iJ^as||dE**'EtSas
+_«	kom'iJ^as||iTkJ^'ERDas
+_»	kom'iJ^as||dE**'EtSas
 
-^	TiRkumfl'Exo
+^	TiRkumfl'exo
 €	eU*o
 %	pOrTj'Ento
 &	ampERs'ant
-@	a*R'OBa
+@	a*R'oBa
 /	ba*Ra
-©	kOpi*R'aIt
+©	kopi*R'aIt
 £	liB@-*as
 ¶	p'a*Rafo
 §	sEkTj'On
 
 
 // numbers
-_0	TE*o
+_0	Te*o
 _1	'uno
 _2	d'Os
 _3	t@-*'es
 _4	kw'at@-*o
 _5	T'inko
 _6	s'eIs
-_7	sj'Ete
+_7	sj'ete
 _8	'OtSo
-_9	nw'Eve
+_9	nw'eve
 _1X	dj'ET
 _11	'OnTe
-_12	d'OTe
-_13	t@-*'ETe
+_12	d'oTe
+_13	t@-*'eTe
 _14	kat'ORTe
 _15	k'inTe
 _20	v'eInte
@@ -78,21 +78,25 @@ _6X	sEs'Enta
 _7X	sEt'Enta
 _8X	OtS'Enta
 _9X	nOv'Enta
-_0C	T'ientOs
+_0C	Tj'EntOs
 _1C0	T'ien         // exactly one hundred
 _1C	T'iento
 _5C	kinj'EntOs
-_7C	s,EtETj'EntOs
-_9C	n,OvETj'EntOs
+_7C	s,eteTj'EntOs
+_9C	n,OveTj'EntOs
 _0M1	m'il
 _1M1	m'il          // no '1' before thousand
-_0M2	mil^'Ones
-_1M2	'unmil^'On
-_0M4	_bil^'Ones
-_1M4	'unbil^'On
+_0M2	miJ^'ones
+_1M2	'unmiJ^'On
+_0M4	_biJ^'onEs
+_1M4	'unbiJ^'On
+?2 _0M2	mij:'ones
+?2 _1M2	'unmij:'On
+?2 _0M4	_bij:'onEs
+?2 _1M4	'unbij:'On
 _0and	i
-_dpt	kOma
-//_roman	ROm'ano
+_dpt	koma
+//_roman	Rom'ano
 
 vi	vi	// not a Roman number
 

dictsource/es_rules

@@ -5,6 +5,7 @@
 // ?1  Castilian
 // ?2  Latin America
 
+.L01 j w l r d g n m
 
 .group a
        _) a (_       a
@@ -12,6 +13,8 @@
           ai         aI
           ay (K      aI
           ay (_      'aI
+          au         aU
+          au (_      'aU
 
 
 .group b
@@ -20,7 +23,9 @@
        _) b          b
        m) b          b
        n) b          b
-
+          b (L01     b
+          b (iA      b
+          b (uA      b
 
 .group c
        _) c (_       Te
@@ -39,12 +44,13 @@
 
 .group e
        _) e (_       e
-          e          E
-          e (_       e
+          e          e
+          e (CK      E
           ei         eI
           ey (K      eI
           ey (_      'eI
-       _) eu         eU
+          eu         eU
+          eu (_      'eU
 
 
 .group f
@@ -89,6 +95,7 @@
        _) l (_       Ele
           l          l
           ll         J^
+?2     A) ll (A      j:
 
 
 .group m
@@ -109,8 +116,9 @@
 
 .group o
        _) o (_       o
-          o          O
-          o (_       o
+          o          o
+          o (CK      O
+          oi         oI
           oy (K      oI
           oy (_      'oI
 
@@ -151,6 +159,7 @@
        _) u (_       u
           u          u
           u (A       w
+          u (y_      w
        l) u (A       %u
        r) u (A       %u
 
@@ -163,8 +172,15 @@
        _) v          v#
 ?1     m) v          b
 ?1     n) v          b
+?1        v (L01     b
+?1        v (iA      b
+?1        v (uA      b
 ?2        v          v#
 ?2     _) v          v
+?2        v (L01     v
+?2        v (j       v
+?2        v (iA      v
+?2        v (uA      v
 
 
 .group w
@@ -188,6 +204,7 @@
        n) y          J
       _l) y          J
           y (A       J^
+?2     A) y (A       j:
 
 
 .group z
@@ -211,7 +228,7 @@
       _)  ++ (_      masm'as
  \+\+\+)  +          // ignore + after the first 3
 
-          #          almOaD'il^a
+          #          almOaD'iJ^a
       \#) #
 
      __)  - (_D      m'EnOs

dictsource/fr_list

@@ -5,7 +5,7 @@
 
 // 2006-11-18  Gilles Casse  <gcasse@oralux.org>
 //
-// Updated 2008-02-20 Michel Such  <michel.such@free.fr>
+// Updated 2008-02-24 Michel Such  <michel.such@free.fr>
 //
 //	* Numbers, a few abbreviations and exceptions.
 //
@@ -32,7 +32,7 @@ e	@
 f	Ef
 g	Ze
 h	aS
-// i	i
+i	i
 ï	i:tRema
 j	Zi
 k	ka
@@ -269,11 +269,12 @@ nous	$u+ $verbf
 vous	$u+ $verbf
 elles	$u+ $verbf
 ils	$u+ $verbf
-on	$u $verbf
+on	O~n2 $u $verbf
 me	$u $verbf
 te	$u $verbf
 se	$u $verbf
 lui	$u $verbf
+ça	$u $verbf
 
 ce	$u $nounf
 cette	$u $nounf
@@ -368,44 +369,29 @@ tout	t'ut2 $u
 // Letters which can be words
 //===========================
 à       a:aksA~gRav  $atend
-i       i      $atend
 y       i:gR'Ek      $atend
 
 
 
 // pronunciation exceptions
 
+aspic          aspik
 consent          kO~s'A~t2
 scient           si
-bénéficient      benefisi $verb
 coing     kwE~
-concurrent       kO~kyR'A~
-content          kO~t@-t2 $verb
+convent       kO~vA~
 couvent          k'uvt2  $verb
 (couvent couvent)         kuvA~||k'uvt2
 dessus          d@sy
 dessous          d@su
 divers          divErz2
-►évident          evid'A~
-évident          evidt2 $verb
-ferment           fErm'A~
-ferment           f'Ermt2  $verb
-firent           f'irt2
 parent            paR'A~
 parent            p'art2  $verb
 pastis     pastis
 poing     pwE~
-président        pRezid'A~
-président        pRezidt2 $verb
-résident         Rezid'A~
-résident         Rezidt2 $verb
 ressent          r@s'A~t2
-récurrent        RekyR'A~
 riz       ri
-sergent           sErZ'A~
-serment           sErm'A~
-serpent           sErp'A~
-torrent           tOR'A~
+sphincter           sfE~ktEr
 tunis           tynis
 
 
@@ -421,14 +407,10 @@ montréal         mO~Real
 (cent une)       s'A~||yn
 (cent onzième)   s'A~||O~zj'Em
 (cent onze)      s'A~||O~z
-(on habille)     O~||nab'ij
 absent        absA~
-accident      aksidA~
-adéquat       adekuat2
-adéquate      adekuat
 airbus	      Erbys
 anus     anys
-ardent        aRdA~
+auvent     ovA~
 bissus    bisys
 bonus     bonys
 bus	      bys
@@ -437,27 +419,22 @@ campus   kA~pys
 casus    kazys
 choeur        k@r
 cocus     coky
-discident     disidA~
 estomac     Estoma
 fils	      fis
-imprudent     E~pRydA~
 iris	      iris
 juin	      ZyE~
 laser	      lazEr
 malus      malys
 mars	      maRs
 minus      minys
-occident      OksidA~
 orient        ORjA~
+paravent        paravA~
 pays	      pEi
 phallus    falys
 polder    pOldEr
-prudent       pRydA~
-secret        sYkRE
 souris       suri
 sus       sys
 sus       sy $verb
-to       tu
 vénus       venys
 
 
@@ -541,23 +518,16 @@ flirt		fl@Rt
 // words from other languages
 about		_^_en
 acer             asEr
-alone           _^_en
-also            _^_en
 and             _^_en
 amazon		amaz'On
 apple		_^_en
-april	_^_en
 at		_^_en
 (audible manager) odibl||manadZ@r
-be		_^_en
 bit		_^_en
 bluetooth 		blut'us
-both 		_^_en
-can		_^_en
 computer		_^_en
 (cyber link) 		_^_en
 debian		dEbjAn
-don't		_^_en
 driver		_^_en
 emacs		Emaks
 emacspeak	Emakspi:k
@@ -565,13 +535,10 @@ end             _^_en
 epson		EpsOn
 espeak		@spi:k
 ethernet	etERnEt
-ever	_^_en
 exit	egzit
 eye	      _^_en
 eyes	      _^_en
-false	_^_en
 (file zilla)		fajl||zija
-for	      _^_en
 get		_^_en
 google		gu:g@l
 gnome		gnom
@@ -580,19 +547,10 @@ hamburger		_!A~b@rg@r
 help		_^_en
 hot		_^_en
 in		_^_en
-into		_^_en
 inside		_^_en
-insight		_^_en
 internet	E~tERnEt
 (internet explorer) E~tErn'Et||EksplOr'@r
-is		_^_en
-it		_^_en
-its		_^_en
-june	_^_en
-july	_^_en
 klaxon          klaksOn
-later    		_^_en
-latest    		_^_en
 layer    		_^_en
 let 		_^_en
 liszt 		list
@@ -601,46 +559,25 @@ login 		_^_en
 logon 		_^_en
 (mac os x)	makoEsiks
 made	_^_en
-mail	_^_en
-march	_^_en
 media     		medja
 messenger 		_^_en
 my		_^_en
-name		_^_en
-never	_^_en
 not		_^_en
 null	_^_en
-october		_^_en
-of         _^_en
-often      _^_en
 ok         oke
 open            Op'@n
 outlook         autluk
-outside		_^_en
-outsider		_^_en
 paint    		_^_en
+paper    		_^_en
 player    		_^_en
 redhat		REdat
 same		_^_en
 sametime	_^_en
-she	_^_en
 schubert	SubER
-since 		_^_en
 sun		_^_en
-sure		_^_en
 (text aloud) tEkst||@lawd
-their	_^_en
-there	_^_en
-these	_^_en
-those	_^_en
-they	_^_en
-thus	_^_en
-true	_^_en
 ubuntu		ubuntu
 up		_^_en
-viking		vikiN
-was		_^_en
-won't		_^_en
 yacht		jot
 yes		_^_en
 

dictsource/fr_rules

@@ -4,7 +4,7 @@
 
 // 2006-11-18  Gilles Casse  <gcasse@oralux.org>
 //
-// Updated: 2008-02-20 Michel Such  <michel.such@free.fr>
+// Updated: 2008-02-24 Michel Such  <michel.such@free.fr>
 //
 //	* The rules are based on Cicero TTS.
 //         Y
@@ -15,9 +15,6 @@
 // Y  front vowels: e i y é ê è î
 // K  not a vowel  (i.e. consonant, space, )
 
-.L01 a b c d e f g h i l p q r t v
-.L02 a e i o u y
-
 
 .group a
 		ae (_       e		// reggae vitae
@@ -53,19 +50,7 @@
 		a           a		// bateau
 
 // group a: English section
-	_C)	ad  (_        _^_en		// bad, had, sad
-	_)	again  (X        _^_en		// again, against
-		alk       _^_en		// talk, walk
-	f)	all (en_       _^_en		// fallen
 	sm)	all       _^_en		// small, smaller
-		ank       _^_en		// tank, blank
-		ark       _^_en		// dark, park
-		ainl       _^_en
-		ainm       _^_en
-	_C)	ast (_      _^_en
-	_C)	ast (eX      _^_en
-		ather (_       _^_en		// rather
-		aunc       _^_en
 		a (wC      _^_en
 
 
@@ -78,14 +63,10 @@
 		back       _^_en
 		bbl       _^_en
 		bird       _^_en		// bird
-	_)	blue (X      _^_en		// probably
-		bly (_      _^_en		// probably
 		board      _^_en              // board, keyboard
 		box (_      _^_en              // box, inbox, outbox
 		bug (_      _^_en              // bug, debug
 		bug (g      _^_en              // debugger
-		buil       _^_en
-	_)	buy       _^_en
 	_)	by       _^_en              // by, bye, bypass
 
 
@@ -127,15 +108,14 @@
 		c           k		// recoin donc
 
 	s)	cien (t	    jA~		// conscient scientifique
-		cien (t_    i		// ils appr�ient, remercient.
-	i)	cien (t_    sjA~	// coefficient
+		cien (t_    si		// ils appr�ient, remercient.
+	effi)	cien (t_    sjA~	// coefficient
+	défi)	cien (t_    sjA~	// déficient
 
 // group c: English section
 		cast (_      _^_en		// broadcast
-		chme       _^_en
-		cult (_      _^_en		// difficult
+	C)	ch (_      _^_en		// french
 		cut (_      _^_en		// cut, shortcut
-		cy (_      _^_en		// currency, frequency
 
 
 .group d
@@ -162,18 +142,9 @@
 		d           d		// don bled
 
 // group d: English section
-	')	d (_      _^_en
 		day (_        _^_en
-	_)	dec        _^_en
-		dece        _^_en
-	_)	def        _^_en
-		dele        _^_en
-		deter        _^_en
 	_)	devic        _^_en
-		diffe        _^_en
 		dle (X        _^_en // bundle, handle
-		dly (_        _^_en
-		dy (_        _^_en
 
 
 .group e
@@ -218,11 +189,58 @@
 	Ci)	en (nA      E		// mienne
 
 	éC)	en (t_      A~		// récent différent élément
+	étic)	en (t_      A~		// réticent
 
 	_C)	en (t_      A~		// cent vent lent dent
 	cc)	en (t_      A~		// accent
+	Asc)	en (t_      A~		// luminescent
+	dol)	en (t_      A~		// dolent
+	imCoC)	en (t_      A~		// impotent
+	inCoC)	en (t_      A~		// innocent, indolent
 
-	_jac)	en (t_      A~		// sous-jacent
+	XACcid)	en (t_      A~		// accident, occident
+
+	_ag)	en (t_      A~		// agent
+	_arC)	en (t_      A~		// argent, ardent, arpent
+	_urg)	en (t_      A~		// urgent
+	V_urg)	en (t_      		// urgent (verbe)
+
+	_émin)	en (t_      A~		// éminent
+	immin)	en (t_      A~		// imminent
+	oémin)	en (t_      A~		// proéminent
+	jac)	en (t_      A~		// sous-jacent, adjacent
+	_lat)	en (t_      A~		// latent
+	lig)	en (t_      A~		// intelligent
+	man)	en (t_      A~		// permanent
+	ndig)	en (t_      A~		// indigent
+	_pat)	en (t_      A~		// patent
+	rmam)	en (t_      A~		// firmament
+	xig)	en (t_      A~		// exigent
+	Vxig)	en (t_      		// exigent
+	éCerg)	en (t_      A~		// détergent
+	V_éCerg)	en (t_      		// émergent
+	verg)	en (t_      A~		// convergent, divergent
+	Vverg)	en (t_      		// convergent, divergent (verbe)
+
+	_serg)	en (t_      A~		// sergent
+	CACCim)	en (t_      A~		// condiment, gentiment
+	dim)	en (t_      A~		// rudiment
+	Agim)	en (t_      A~		// régiment
+	Alim)	en (t_      A~		// poliment
+	inim)	en (t_      A~		// infiniment
+	manim)	en (t_      A~		// maniment
+	onim)	en (t_      A~		// boniment
+	plim)	en (t_      A~		// compliment
+	Atim)	en (t_      A~		// bâtiment
+	_cim)	en (t_      A~		// ciment
+	_pim)	en (t_      A~		// piment
+	erm)	en (t_      A~		// ferment, serment
+	Vferm)	en (t_      		// ferment, referment (verbe)
+	Arp)	en (t_      A~		// arpent, serpent
+	CArr)	en (t_      A~		// conncurrent, torrent
+	mitt)	en (t_      A~		// intermittent
+	énit)	en (t_      A~		// pénitent
+	tourm)	en (t_      A~		// tourment
 
 	ti)	en (t_      E~		// retient
 	ati)	en (t_      A~		// patient
@@ -239,18 +257,22 @@
 	mom)	en (t_      A~		// moment
 	Aaim)	en (t_      A~
 	Caim)	en (t_      A~
-	cum)	en (t_      A~
+	dum)	en (t_      A~
 	gum)	en (t_      A~
-	lum)	en (t_      A~
-	rum)	en (t_      A~		// prudemment
+	oCum)	en (t_      A~
+	rum)	en (t_      A~
 
 	_cli)	en (t      A~		// client
+	éCid)	en (t_      A~
+	VéCid)	en (t_
+	Xtrid)	en (t_      A~
 	_laur)	en (t_      A~
 	_mécont) en (t_     A~
 	_cont)	en (t_      A~
-	mpét)	en (t_      A~		// compétent
-	prés)	en (t_      A~
+	_Vcont)	en (t_      t2
+	éC)	en (t_      A~		// compétent
 	_souv)	en (t_      A~		// souvent
+	Cud)	en (t_      A~
 
 	s_couv)	en (t_      		// elles couvent
 	qui_couv) en (t_
@@ -271,7 +293,6 @@
 		en (s_      A~
 		en (CA      A~      	// pentathlon
 		en (CC      A~      	// entre
-		en (ch_      En      	// french
 	sp)	ens (_      Ens  // suspens
 	éC)	ens (_      A~z2  // dépens démens
 
@@ -386,6 +407,8 @@
 	tr) 	e (CrA      @  		// entreprise
 
  		e (CC       E  		// infect pelle mettre
+ 		e (CC       E  		// infect pelle mettre
+ _s)		e (cr       @  		// secret
  		e (C_       E
  		es (_       z2
 
@@ -403,37 +426,21 @@
 // group e: English section
 		eac		_^_en
 	C)	ead		_^_en      		// read, head
-		eaf       _^_en		// leaf
-		eag       _^_en		// eagle
 		eak           _^_en		// break, speak
 	Cr)	eam		_^_en      		// stream
 	Cl)	ean		_^_en      		// clean
-	l)	ean		_^_en      		// lean
-	m)	ean		_^_en      		// mean
 		ear		_^_en      		// ear, search
 		eas  	_^_en      		// please
-		eat  	_^_en      		// eat, seat
 		eave       _^_en		// leave
 		ed (_  	_^_en
 	_)	edit   	_^_en
 		ee          _^_en		// meeting
-		eft (_     _^_en		// left
-		ehen       _^_en		// comprehensive
 		eing       _^_en		// being, goeing
 		eive       _^_en		// receive
-		eith       _^_en		// either neither
 		ej       _^_en
-	_) 	el (AC        _^_en		// element, eliminate
 	 	eld (_        _^_en		// field
-	 	elf (_        _^_en		// shelf
-	 	elves (_        _^_en		// shelves
-		ember (_       _^_en
 	 	enter (_        _^_en		// enter, center
-	 	ely (_        _^_en		// lately
 	_) 	enhan         _^_en
-	ooC)	er (_       _^_en		// scooter
-		ern (_       _^_en		// western
-	_C)	etter (_      _^_en     	// letter, better
 	ack)	et (_      _^_en     	// racket, packet
 	ock)	et (_      _^_en     	// rocket, pocket
 	qu) 	est (_        _^_en		// request
@@ -454,16 +461,9 @@
 	_neu)	f (_heures  v
 
 // group f: English section
-		fail (_       _^_en  // fail
-		fail (A       _^_en  // failure
-		faith        _^_en  // failure
 	_)	fire       _^_en
 		fly        _^_en
-		ford (_        _^_en
 	_)	frame         _^_en
-	_)	freq         _^_en
-		friend        _^_en		// friend, friendly
-		fy (_     _^_en
 
 
 .group g
@@ -512,7 +512,6 @@
 	@@)	gate (X         _^_en
 		gh          _^_en		// high, higher
 		girl        _^_en		// girl
-		gly (_         _^_en
 		give (X         _^_en
 		gy (_         _^_en
 
@@ -599,7 +598,6 @@ _)		h	(umo			// humour
 
 // group h: English section
 _)		half  	_^_en
-_)		h (As_	_^_en		// has, his
 		here (_	_^_en		// here
 
 
@@ -631,30 +629,12 @@ _)		h (As_	_^_en		// has, his
                 ique (_	    ik
 
 // group i: English section
-	A_)	i (_        _^_en
-	C_)	i (_        _^_en
-	_)	i (_A        _^_en
-	_)	i (_C        _^_en
-	L02C)	ic (_        _^_en
-	L02CC)	ic (_        _^_en
-	@C)	id (_        _^_en
-		idd         _^_en
 		iev         _^_en
 		ife (_      _^_en
-		igg         _^_en
-		ike (X        _^_en		// bike, like
-		ild         _^_en
-		ilt         _^_en
 		ing (_      _^_en		// parking
 		ind (_     _^_en		// find, mind
 	_C)	ind (er_     _^_en		// finder, reminder
 	_AC)	ind (er_     _^_en		// finder, reminder
-		ink      _^_en		// link, pink
-		ip (_       _^_en		// chip, ship
-		ious (_       _^_en		// various
-		iously (_       _^_en		// previously
-		ist (_       _^_en		// tourist
-		ism (_       _^_en		// tourism
 		ize          _^_en		// realize
 
 
@@ -667,7 +647,8 @@ _)		h (As_	_^_en		// has, his
 		ïn (C       E~		// coïncider
 		ïn (_       E~
 		ïs        is  // maïs, archaïsme
-		ï (q       i  // archaïquee
+		ï (q       i  // archaïque
+		ï (c       i  // laïc
 	a) 	ï           j		// aïeul
 		ï           i		// ambiguïté
 
@@ -675,12 +656,16 @@ _)		h (As_	_^_en		// has, his
 .group j
 		j           Z		// adjoint joujoux
 
+// group j: English section
+		ject (_           _^_en
+
 
 .group k
 		k           k		// kafka
 
 
 // group k: English section
+	AC)	k      _^_en		// blank, black, dark
 		ke (X     _^_en		// basket, make, take
 		key     _^_en
 		ky (_     _^_en
@@ -694,15 +679,8 @@ _)		h (As_	_^_en		// has, his
 	ui)	ll          j		// juillet
 
 // group l: English section
-	')	ld (_      _^_en
-	')	ll (_      _^_en
 		less (_      _^_en  // noiseless
 	_)	live       _^_en
-		ll (_      _^_en
-		lly (_      _^_en
-		lk (_      _^_en
-		lessly (_      _^_en  // endlessly
-		lord (_        _^_en
 
 
 .group m
@@ -710,9 +688,7 @@ _)		h (As_	_^_en		// has, his
 		mm          m		// pomme
 
 // group m: English section
-	')	m (_         _^_en
 		mail (A       _^_en  // mailer
-		may        _^_en
 		mov (A         _^_en		// move, movy
 
 
@@ -726,10 +702,6 @@ _)		h (As_	_^_en		// has, his
 	A) 	ng (_       N		// parking meeting
 		nn          n		// panne
 
-// group n: English section
-		n't (_       _^_en
-		nunc       _^_en
-		ny (_       _^_en
 
 
 .group o
@@ -808,22 +780,14 @@ _)		h (As_	_^_en		// has, his
 	 	oa (X       _^_en		// approach, load
 	 	oa (CeX       _^_en
 	 	oach       _^_en
-	 	oes (_      _^_en
 		old (er_       _^_en		// folder, older
 C)		oing (_      _^_en	// going
-	_aC)	ong (_       _^_en		// along, among
 		oo  (Ce     _^_en		// boomer
 		oo (X       _^_en		// pool
 	C)	oot        _^_en		// bootable football
 	_)	one       _^_en
-	_)	onl       _^_en
-		orm (_      _^_en
 	_sC)	ot       _^_en
-		othe       _^_en		// other, mother
-	C) 	ou (ld_      _^_en		// could, should
 	 	oun (C      _^_en		// bounce, found
-	_C) 	ouse       _^_en		// mouse, house
-	C) 	outh       _^_en		// mouth, south
 	_)	over        _^_en		// over
 		ow          _^_en		// cow, town, down
 		oy (X       _^_en		// boy, toy
@@ -859,16 +823,8 @@ C)		oing (_      _^_en	// going
 // group p: English section
 		pad (_        _^_en
 		plug 	_^_en		// plug
-		pmen 	_^_en
-		ply (_        _^_en		// simply
-	_) pre	(L01 	_^_en
-	 press	(_ 	_^_en
 		print (_        _^_en		// print
 		printer (_        _^_en		// printer
-	_)	prove (X       _^_en		// prove
-	_AC)	prove (X       _^_en		// improve
-	_AC)	provem       _^_en		// improvement
-		py (_        _^_en		// copy
 
 
 .group q
@@ -879,6 +835,7 @@ C)		oing (_      _^_en	// going
  	_A)	qu (ilat      ky		// équilatéral
  	_C)	qu (a      kw		// squale square
  	_A)	qu (a      kw		// équateur
+ 	dA)	qu (a      kw		// adequate
  	_)	qu (artz      kw		// quartz
 		qu          k		// quatre
  	_)	que (_      k@		// que
@@ -901,8 +858,6 @@ mou)	rr          RR		// mourrai
 	_)	real         _^_en
 		rese         _^_en
 		rst (_        _^_en
-	_)	rule (X        _^_en		// rule
-		ry (_        _^_en		// theory
 
 
 .group s
@@ -978,20 +933,13 @@ e)		s (_h      z2
 		s (v         z
 
 // group s: English section
-	')	s (_         _^_en
 		sh (_         _^_en
-		she (C_         _^_en
 		ship          _^_en		// friendship
 		shop          _^_en		// shop
 		sh  (At          _^_en		// shut, shot
-		shout          _^_en
 		sk         _^_en
 	_)	smil        _^_en   // smile, smiley
 		spy         _^_en
-	A)	ss (_       _^_en		// boss, cross
-	_)	state        _^_en		// state, statement
-	C) 	s (ton        _^_en		// winston
-	 	stone (_        _^_en
 		sy (_      _^_en
 
 
@@ -1057,8 +1005,6 @@ C)		te (_      t@-  		// patte
 	_ce) 	t (_        t
 
 // group t: English section
-		tch (_      _^_en
-	_)	th (AX      _^_en      // that, this, then, than
 	C)	th (_      _^_en      // month
 	_)	time      _^_en
 		tle (_     _^_en
@@ -1067,10 +1013,7 @@ C)		te (_      t@-  		// patte
 		try (_     _^_en  // try, country
 	sCar)	t (_        _^_en	// smart, start
 	sCar)	t (er_        _^_en	// starter
-		tme         _^_en
-		tne         _^_en
 	_)	tun (AX        _^_en
-		ty (_     _^_en  // party, buty
 
 
 .group u
@@ -1104,28 +1047,21 @@ C)		te (_      t@-  		// patte
 	g) 	u (ë        y	// ambiguë
 
 // group u: English section
-	C)	u  (ch_        _^_en  // much such
 	C)	ui  (ce        _^_en  // produice, juice
 		umber (_       _^_en
 		ump (_       _^_en
 		unch          _^_en
-	_C)	unct          _^_en  // function, punctuation
 		under          _^_en  // under, understand, thunder
 	_)	up  (C        _^_en  // upper, update
-		upt (_       _^_en
 	up  (_        _^_en  // setup
-	C)	ur  (ch        _^_en  // church
 		urn (X        _^_en  // burn, turn
 		ust (_        _^_en  // just, trust
-	rib)	u  (teX        _^_en  // tribute, attribute
 
 
 .group v
 		v           v
 
 // goup v: English section
-	')	ve (_      _^_en
-		vail  (A       _^_en
 		voice  (X       _^_en
 		void  (_       _^_en
 		vy  (_       _^_en
@@ -1142,11 +1078,9 @@ C)		te (_      t@-  		// patte
 		wr 	  _^_en    // write
 		w (ACC 	  _^_en    // wash, wish, with
 		way 	  _^_en    // way, away
-		wh 	  _^_en    // what, which, who
 	_)	wi 	  _^_en    // wire
 		win 	  _^_en    // winner, window
 		wise 	  _^_en
-	_)	wom 	  _^_en    // woman
 		wor 	  _^_en    // word, world
 
 
@@ -1177,9 +1111,6 @@ C)		te (_      t@-  		// patte
 	si) 	x (iè   z	// sixième
 	deu) 	x (iè   z	// deuxième
 
-// group x: English section
-		xamp		_^_en
-
 
 .group y
 		y           i	// cryogénique myope

dictsource/ta_rules

@@ -55,7 +55,7 @@
 	க (் 	g
      ்) க	ga
      ்) க (B	g
-     _) க	ka
+     _) க	kV
      _) க (B	k
 	க்க	k:a
 	க்க (B	k:
@@ -78,7 +78,7 @@
 
 	ட	d.a
 	ட (B	d.
-     _) ட	t.a
+     _) ட	t.V
      _) ட (B	t.
 	ட்ட	t.a
 	ட்ட (B	t.
@@ -88,7 +88,7 @@
 
 	த	da
 	த (B	d
-     _) த	ta
+     _) த	tV
      _) த (B	t
 	த்த	t:a
 	த்த (B	t:
@@ -101,10 +101,12 @@
 
 	ப	ba
 	ப (B	b
-     _) ப	pa
+     _) ப	pV
      _) ப (B	p
 	ப்ப	p:a
 	ப்ப (B	p:
+     ஃ) ப	fa
+     ஃ) ப (B	f
 
 	ம	ma
 	ம (B	m
@@ -117,7 +119,7 @@
 
 	ற	Ra
 	ற (B	R
-	ற் (ற	t	// RR -> tR
+	ற் (ற	t.	// RR -> t.R
 
 	ல	la
 	ல (B	l

dictsource/tr_list

@@ -9,8 +9,8 @@ c	dZE
 d	dE
 e	E
 f	fE
-g	g;E
-ğ	jumuS'ak||g;'E
+g	JE
+ğ	jumuS'ak||J'E
 h	hE
 ı	u#
 i	i
@@ -19,7 +19,7 @@ k	ka
 l	lE
 m	mE
 n	nE
-o	O
+_o	O
 ö	W
 p	pE
 q	kvE
@@ -37,41 +37,79 @@ z	zE
 
 _??	sEmb'Ol
 
-_0	su#fu#r
-_1	bir
-_2	iki
+_0	su#fu#R
+_1	biR
+_2	ici
 _3	YtS
-_4	dWrt
+_4	dWRt
 _5	beS
 _6	altu#
 _7	jedi
-_8	sekiz
+_8	seciz
 _9	dokuz
 _10	on
-_11	'onbir
-_12	'oniki
+_11	'onbiR
+_12	'onici
 _1X	on
-_2X	jirmi
+_2X	jiRmi
 _3X	otuz
-_4X	ku#rk
+_4X	ku#Rk
 _5X	elli
 _6X	altmu#S
 _7X	jetmiS
 _8X	seksEn
 _9X	doksan
 _0C	jyz
-_2C	'ikijyz
+_2C	'icijyz
 _0M1	bIn
 _1M1	bIn
 _0M2	miljon
-_1M2	miljon
 _0M3	miljar
-_1M3	miljar
-_dpt	_virg,Yl_|
+_dpt	_viRg,Yl_|
+
+
+// function words
+
+// conjunctions
+ve	$brk	// and
+yoksa	$brk	// or
+veya	$brk	// or
+ama	$brk $1	// but
+ki	$brk	// that
+
+mı	$u	// question
+mi	$u
+mu	$u
+mü	$u
+
+
+// pronouns
+bu	$u+	// this
+şu	$u+	// that
+
+ben	$u+	// I
+sen	$u+	// you
+o	$u+	// it
+siz	$u+	// we
+biz	$u+	// you
 
 
 
-// exceptions
+bir	$u+	// a (or one)
 
+da	$u	// also
+de	$u
+ta	$u
+te	$u
+
+ile	$u
+
+
+// EXCEPTIONS
+
+// Person Names
+
+// Place Names
 ankara		$1
+istanbul	$2
 

dictsource/tr_rules

@@ -2,11 +2,27 @@
 // Turkish translation rules
 // This file is UTF-8 encoded
 
+// Stress rule: Right-most vowel, but stop before the vowel which is
+// marked as unstressed [%].
+
+// unvoiced consonants
+.L01 ç f h k p q s ş t 
+
+.replace
+`	'
 
 .group a
         a        a
 
-        avru (pa avr'u
+     @) a (_S1   a
+
+	acak (_S3  adZak
+	acakmış (_S7   adZakm%u#S
+	ayım (_S4  aju#m
+	alım (_S4  alu#m
+	asın (_S4  asu#n
+
+	avru (pa avr'u
 
 .group b
         b        b
@@ -20,49 +36,149 @@
 .group d
         d        d
 
+	da (_S2  da
+	de (_S2  de
+	dan (_S3  dan
+	den (_S3  den
+
+        // declare these suffices so that preceding "ma/me" is unstressed
+	dim (_S3   dim
+	dik (_S3   dik
+	din (_S3   din
+	diniz (_S5 diniz
+	di (_S2    di
+	diler (_S5 dileR
+	dım (_S3   du#m
+	dık (_S3   du#k
+	dın (_S3   du#n
+	dınız (_S5 du#nu#z
+	dı (_S2    du#
+	dılar (_S5 du#laR
+	dum (_S3   dum
+	duk (_S3   duk
+	dun (_S3   dun
+	dunuz (_S5 dunuz
+	du (_S2    du
+	dular (_S5 dular
+	düm (_S3   dym
+	dük (_S3   dyk
+	dün (_S3   dyn
+	dünüz (_S5 dynyz
+	dü (_S2    dy
+	düler (_S5 dyler
+
+	dir (_S3  %diR
+	dır (_S3  %du#R
+	dur (_S3  %duR
+	dür (_S3  %dyR
+
      _) d (eğil  'd
 
+
 .group e
         e        e
 
+     @) e (_S1   e
+
+	ecek (_S4  edZek
+	ecekmiş (_S7   edZekm%iS
+	eyim (_S4  ejim
+	elim (_S4  elim
+	esin (_S4  esin
+
+
 .group f
         f        f
 
 .group g
         g        g
+        g (i     J
+        g (e     J
+        g (ü     J
+        g (ö     J
 
 .group ğ
         ğ        Q
-     A) ğ        :
+     A) ğ        :||
+     i) ğ        j
+     e) ğ        j
 
 .group h
         h        h
 
+
 .group i
         i        i
 
+     @) i (_S1	i
+
+	im (_S2  %im
+	iniz (_S4  %iniz
+	iz (_S2  %iz
+	iyor (_S4  ij%oR
+	in (_S2  in
+
+
 .group ı
         ı        u#
 
+     @) ı (_S1   u#
+
+	ım (_S2  %u#m
+	ınız (_S2  %u#nu#z
+	ız (_S2  %u#z
+	ıyor (_S4  u#j%oR
+	ın (_S2  u#n
+
+
 .group j
         j        Z
 
 .group k
         k        k
+        k (i     c
+        k (e     c
+        k (ü     c
+        k (ö     c
+
+	ken (_S3  %cen
+
 
 .group l
         l        l
 
+	la (_S2  %la
+	le (_S2  %le
+
+	lar (_S3  laR
+	ler (_S3  leR
+
+	li (_S2  li
+	lı (_S2  lu#
+	lu (_S2  lu
+	lü (_S2  ly
+
+	lik (_S3  lik
+	lık (_S3  lu#k
+	luk (_S3  luk
+	lük (_S3  lyk
+
+
 .group m
         m        m
 
-        ma (_S2  m%a
-        me (_S2  m%e
-        mı (_S2  m%u#
+	ma (_S2  %ma
+	me (_S2  %me
+
 
 .group n
         n        n
 
+	nin (_S3  nin
+	nın (_S3  nu#n
+	nun (_S3  nun
+	nün (_S3  nyn
+
 .group o
         o        o
 
@@ -76,23 +192,76 @@
         q        k
 
 .group r
-        r        r
+        r        R
+     A) r (A     *
+
+	ra (_S2  %Ra
+	re (_S2  %Re
+	ra (_N   Ra    // only an unstressed suffix if another suffix follows
+	re (_N   Re
+
 
 .group s
         s        s
 
+	sınız (_S5  %su#nu#z
+	siniz (_S5  %siniz
+	sunuz (_S5  %sunuz
+	sünüz (_S5  %synyz
+
+	sam (_S3   %sam   // if
+	sak (_S3   %sak
+	san (_S3   %san
+	sanız (_S5 %sanu#z
+	sa (_S2    %sa
+	sem (_S3   %sem
+	sek (_S3   %sek
+	sen (_S3   %sen
+	seniz (_S5 %seniz
+	se (_S2    %se
+
+
 .group ş
         ş        S
 
 .group t
         t        t
 
+	ta (_S2  ta
+	te (_S2  te
+	tan (_S3  tan
+	ten (_S3  ten
+
+   L01) tir (_S3  %tiR
+   L01) tır (_S3  %tu#R
+   L01) tur (_S3  %tuR
+   L01) tür (_S3  %tyR
+
+
 .group u
         u        u
 
+     @) u (_S1   u
+
+	um (_S2  %um
+	unuz (_S4  %unuz
+	uz (_S2  %uz
+	uyor (_S4  uj%oR
+	un (_S2  un
+
+
 .group ü
         ü        y
 
+     @) ü (_S1   y
+
+	üm (_S2  %ym
+	ünüz (_S4  %ynyz
+	üz (_S2  %yz
+	üyor (_S4  yj%oR
+	ün (_S2  yn
+
+
 .group v
         v        v
 
@@ -103,11 +272,55 @@
         x        ks
      K) x        z
 
+
 .group y
         y        j
 
+	ya (_S2   ja
+	ye (_S2   je
+	yi (_S2   ji
+	yı (_S2   ju#
+	yu (_S2   ju
+	yü (_S2   jy
+
+	yim (_S3  %jim
+	yım (_S3  %ju#m
+	yum (_S3  %jum
+	yüm (_S3  %jym
+
+	yiz (_S3  %jiz
+	yız (_S3  %ju#z
+	yuz (_S3  %juz
+	yüz (_S3  %jyz
+
+	yiniz (_S5  %jiniz
+	yınız (_S5  %ju#nu#z
+	yunuz (_S5  %junuz
+	yünüz (_S5  %jynyz
+
+	yla (_S3  %jla
+	yle (_S3  %jle
+
+	yacak (_S5  jadZak
+	yecek (_S5  jedZek
+	yacakmış (_S8   jadZakm%u#S
+	yecekmiş (_S8   jedZekm%iS
+
+	yayım (_S5  jaju#m
+	yalım (_S5  jalu#m
+	yasın (_S5  jasu#n
+	yeyim (_S5  jejim
+	yelim (_S5  jelim
+	yesin (_S5  jesin
+
+	yken (_S4  jc%en
+
+
 .group z
         z        z
 
 .group
         $        dolar
+        ' (Pb    // split a word at ' and translate the first part separately.
+
+

phsource/compile_report

@@ -2,13 +2,13 @@
           new total
     base  99   99
    base2  24  118
-      en  54  148
-    en_n  30  148
-   en_us  34  148
-   en_sc  39  150
-   en_rp  34  148
-   en_wm  30  148
-   en_wi  30  148
+      en  53  147
+    en_n  30  147
+   en_us  34  147
+   en_sc  39  149
+   en_rp  34  147
+   en_wm  30  147
+   en_wi  30  147
       af  38  131
       cy  29  125
       de  33  126
@@ -18,7 +18,7 @@
       fr  44  125
    fr_ca  11  125
       hi  51  136
-      ta  15  138
+      ta  16  139
       hu  23  115
       nl  28  122
       pl  15  110
@@ -46,7 +46,7 @@
       sw  14  108
       th  50  142
       id  14  120
-      tr  14  123
+      tr  18  123
       ku  13  120
       ja   7  104
 
@@ -172,6 +172,7 @@ dzh/dzh_        [dZ] base
                 [dZ;] ru
 dzh/dz_pzd      [J] base
                 [J2] hi
+                [J] tr
 dzh/dz_pzd_     [J] base
                 [J2] hi
 dzh/xdzh        [dZ] base
@@ -657,6 +658,8 @@ ustop/ki        [c] base
                 [c] is
                 [k] zhy
                 [k] sw
+                [J] tr
+                [c] tr
 ustop/kl        [k] base
                 [k] base2
                 [k] en
@@ -1617,8 +1620,7 @@ vowel/u         [u:] en_wi
                 [u] th
                 [u:] th
                 [u] id
-vowel/u#        [u-] en
-                [u:] en_sc
+vowel/u#        [u:] en_sc
                 [Y] tr
 vowel/u_2       [u1] fi
                 [u] sk
@@ -1628,6 +1630,8 @@ vowel/u_2       [u1] fi
 vowel/u#_2      [u-] ta
                 [u-] sv
 vowel/u_3       [yU] ro
+vowel/u#_3      [U] ta
+                [U:] ta
 vowel/u_4       [u:] en_n
 vowel/u#_4      [U] en_sc
                 [u:] en_sc
@@ -1664,7 +1668,6 @@ vowel/uu        [U] en
 vowel/uu#       [U] ku
 vowel/uu_2      [U] base2
                 [U] de
-                [U] ta
                 [U] tr
 vowel/uu_3      [u] af
                 [y] zh

phsource/ph_english

@@ -433,9 +433,3 @@ phoneme W
 endphoneme
 
 
-phoneme u-
-  vowel starttype (u) endtype (u)
-  length 200
-  formants vowel/u#
-endphoneme
-

phsource/ph_spanish

@@ -77,8 +77,8 @@ phoneme v#   // a shorter [v], a little towards [b]
   vowelin  f1=1  f2=1000 -300 -200  f3=-300 100
   vowelout f1=0  f2=1000 -500 -300  f3=-300 60  len=10
   lengthmod 6
-  formants voc/v2+vocw/v%75
-  before _ voc/v_+vocw/v%75
+  formants voc/v2+vocw/v%80
+  before _ voc/v_+vocw/v%80
   switchvoicing f
 endphoneme
 

phsource/ph_tamil

@@ -77,17 +77,23 @@ endphoneme
 
 phoneme U
   vowel  starttype (u)  endtype (u)
-  length 130
-  formants vowel/uu_2
+  length 150
+  formants vowel/u#_3
 endphoneme
 
-
 phoneme u:
   vowel  starttype (u)  endtype (u)
   length 270
   formants vowel/u
 endphoneme
 
+phoneme U:
+  vowel  starttype (u)  endtype (u)
+  length 270
+  formants vowel/u#_3
+endphoneme
+
+
 
 phoneme u-
   vowel starttype (u) endtype (u)

phsource/ph_turkish

@@ -91,3 +91,36 @@ phoneme O
   formants vowel/oo
 endphoneme
 
+
+
+phoneme : //  Lengthen previous vowel by "length"
+  virtual
+  length 180
+endphoneme
+
+
+phoneme c
+  vls pal stop
+  vowelin f1=0  f2=2700 200 500  f3=400 80 len=60
+  vowelout f1=0  f2=2700 300 500  f3=300 80
+  lengthmod 2
+  wave ustop/ki
+  before _ ustop/ki%80
+endphoneme
+
+
+phoneme J
+  vcd pla stop
+  vowelin f1=2  f2=2700 400 600  f3=300 80
+  vowelout f1=2 f2=2700 400 600  f3=300 80 colr=1
+  formants dzh/dz_pzd+ustop/ki%60
+  lengthmod 5
+  switchvoicing c
+endphoneme
+
+
+phoneme l
+import_phoneme l en
+endphoneme
+
+

phsource/phonemes

@@ -993,10 +993,10 @@ endphoneme
 
 phoneme q
   vls uvl stop
-  vowelin f1=1  f2=1700 0 200  f3=-300 80 f4 paus rms=35
-  vowelout f1=1 f2=1700 0 200  f3=-300 80 f4 rms=30
+  vowelin f1=1  f2=1700 0 200  f3=-300 80 paus f4 rms=30
+  vowelout f1=1 f2=1700 -100 200  f3=-300 80  f4 rms=35
   lengthmod 2
-  wave ustop/q%50
+  wave ustop/q%48
   before _ ustop/q%40
 endphoneme
 

src/compiledict.cpp

@@ -810,6 +810,9 @@ void copy_rule_string(char *string, int &state)
 						case 't':
 							sxflags |= SUFX_T;
 							break;
+						case 'b':
+							sxflags |= SUFX_B;
+							break;
 						default:
 							if(isdigit(c))
 								value = (value*10) + (c - '0');

src/dictionary.cpp

@@ -2272,6 +2272,8 @@ int Translator::TranslateRules(char *p_start, char *phonemes, int ph_size, char
 	{
 		c = p_start[ix];
 		word_copy[ix++] = c;
+		if(c == 0)
+			break;
 	}
 	word_copy[ix] = 0;
 
@@ -3096,7 +3098,7 @@ int Translator::RemoveEnding(char *word, int end_type, char *word_copy)
 	word_copy[i] = 0;
 
 	// look for multibyte characters to increase the number of bytes to remove
-	for(len_ending = i = (end_type & 0xf); i>0 ;i--)   // num.of characters of the suffix
+	for(len_ending = i = (end_type & 0x3f); i>0 ;i--)   // num.of characters of the suffix
 	{
 		word_end--;
 		while((*word_end & 0xc0) == 0x80)

src/numbers.cpp

@@ -466,7 +466,7 @@ int Translator::LookupNum3(int value, char *ph_out, int suppress_null, int thous
 
 		Lookup("_0C",ph_100);
 
-		if((hundreds >= 10) && ((langopts.numbers & 0x0800) || (hundreds != 19)))
+		if((hundreds >= 10) && (((langopts.numbers & 0x0800) == 0) || (hundreds != 19)))
 		{
 			ph_digits[0] = 0;
 

src/synthdata.cpp

@@ -35,7 +35,7 @@
 #include "translate.h"
 #include "wave.h"
 
-const char *version_string = "1.31.14  21.Feb.08";
+const char *version_string = "1.31.17  24.Feb.08";
 const int version_phdata  = 0x013105;
 
 int option_device_number = -1;

src/tr_english.cpp

@@ -42,7 +42,7 @@ Translator_English::Translator_English() : Translator()
 	memcpy(stress_lengths,stress_lengths2,sizeof(stress_lengths));
 	langopts.stress_rule = 0;
 
-	langopts.numbers = 0x41 + NUM_ROMAN;
+	langopts.numbers = 0x841 + NUM_ROMAN;
 	langopts.param[LOPT_COMBINE_WORDS] = 2;       // allow "mc" to cmbine with the following word
 }
 

src/tr_languages.cpp

@@ -147,7 +147,7 @@ Translator *SelectTranslator(const char *name)
 			tr->langopts.param[LOPT_PREFIXES] = 1;
 			memcpy(tr->stress_lengths,stress_lengths_de,sizeof(tr->stress_lengths));
 		
-			tr->langopts.numbers = 0x11c19 + NUM_ROMAN;
+			tr->langopts.numbers = 0x11419 + NUM_ROMAN;
 			SetLetterVowel(tr,'y');
 		}
 		break;
@@ -189,7 +189,7 @@ Translator *SelectTranslator(const char *name)
 			tr->langopts.unstressed_wd2 = 2;
 			tr->langopts.param[LOPT_SONORANT_MIN] = 130;  // limit the shortening of sonorants before short vowels
 
-			tr->langopts.numbers = 0xb09;
+			tr->langopts.numbers = 0x309;
 			tr->langopts.numbers2 = 0x2;   // variant form of numbers before thousands
 
 			if(name2 == L_grc)
@@ -218,7 +218,7 @@ Translator *SelectTranslator(const char *name)
 			tr->langopts.unstressed_wd1 = 1;
 			tr->langopts.unstressed_wd2 = 2;
 
-			tr->langopts.numbers = 0x1c09 + NUM_ROMAN;
+			tr->langopts.numbers = 0x1409 + NUM_ROMAN;
 		}
 		break;
 
@@ -258,7 +258,7 @@ Translator *SelectTranslator(const char *name)
 			tr->langopts.param[LOPT_IT_DOUBLING] = 1;
 			tr->langopts.long_stop = 140;
 
-			tr->langopts.numbers = 0x1009;
+			tr->langopts.numbers = 0x1809;
 			SetLetterVowel(tr,'y');
 			tr->langopts.max_initial_consonants = 2;
 			tr->langopts.spelling_stress = 1;
@@ -295,7 +295,7 @@ Translator *SelectTranslator(const char *name)
 
 			tr->langopts.stress_rule = 6;      // stress on last heaviest syllable, excluding final syllable
 			tr->langopts.stress_flags =  0x10004;   // use 'diminished' for unstressed final syllable
-			tr->langopts.numbers = 0x811;
+			tr->langopts.numbers = 0x011;
 			tr->langopts.numbers2 = 0x100;
 			tr->letter_bits_offset = OFFSET_DEVANAGARI;
 
@@ -331,7 +331,7 @@ Translator *SelectTranslator(const char *name)
  			tr->langopts.max_initial_consonants = 5;
 			tr->langopts.spelling_stress = 1;
 
-			tr->langopts.numbers = 0x1c0d + 0x4000 + NUM_ROMAN_UC;
+			tr->langopts.numbers = 0x140d + 0x4000 + NUM_ROMAN_UC;
 			tr->langopts.numbers2 = 0x4a;  // variant numbers before thousands,milliards
 			tr->langopts.replace_chars = replace_cyrillic_latin;
 
@@ -358,7 +358,7 @@ Translator *SelectTranslator(const char *name)
 			tr->langopts.param[LOPT_IT_DOUBLING] = 1;
 			tr->langopts.param[LOPT_COMBINE_WORDS] = 99;  // combine some prepositions with the following word
 
-			tr->langopts.numbers = 0x1809 + NUM_ROMAN;
+			tr->langopts.numbers = 0x1009 + NUM_ROMAN;
 			SetLetterVowel(tr,'y');
 			tr->langopts.spelling_stress = 1;
 SetLengthMods(tr,3);  // all equal
@@ -372,7 +372,7 @@ SetLengthMods(tr,3);  // all equal
 			tr = new Translator();
 			SetupTranslator(tr,stress_lengths_id,stress_amps_id);
 			tr->langopts.stress_rule = 2;
-			tr->langopts.numbers = 0x1809 + NUM_ROMAN;
+			tr->langopts.numbers = 0x1009 + NUM_ROMAN;
 			tr->langopts.stress_flags =  0x6 | 0x10; 
 		}
 		break;
@@ -395,7 +395,7 @@ SetLengthMods(tr,3);  // all equal
 			SetLetterBits(tr,3,"jvr");    // Letter group H
 			tr->letter_groups[1] = is_lettergroup_B;
 			SetLetterVowel(tr,'y');
-			tr->langopts.numbers = 0xe9;
+			tr->langopts.numbers = 0x8e9;
 			tr->langopts.numbers2 = 0x2;
 		}
 		break;
@@ -417,7 +417,7 @@ SetLengthMods(tr,3);  // all equal
 			tr->langopts.param[LOPT_IT_DOUBLING] = 2;    // double the first consonant if the previous word ends in a stressed vowel
 			tr->langopts.param[LOPT_SONORANT_MIN] = 130;  // limit the shortening of sonorants before short vowels
 			tr->langopts.param[LOPT_REDUCE] = 1;        // reduce vowels even if phonemes are specified in it_list
-			tr->langopts.numbers = 0x2709 + 0x800 + NUM_ROMAN;
+			tr->langopts.numbers = 0x2709 + NUM_ROMAN;
 		}
 		break;
 
@@ -451,7 +451,7 @@ SetLengthMods(tr,3);  // all equal
 
 			tr->langopts.stress_rule = 7;   // stress on the last syllable, before any explicitly unstressed syllable
 
-			tr->langopts.numbers = 0x100c69;
+			tr->langopts.numbers = 0x100469;
 			tr->langopts.max_initial_consonants = 2;
 		}
 		break;
@@ -483,7 +483,7 @@ SetLengthMods(tr,3);  // all equal
 			tr->letter_groups[0] = vowels_cyrillic;
 
 			tr->langopts.stress_rule = 4;   // antipenultimate
-			tr->langopts.numbers = 0x0c29 + 0x4000;
+			tr->langopts.numbers = 0x0429 + 0x4000;
 			tr->langopts.numbers2 = 0x8a;  // variant numbers before thousands,milliards
 		}
 		break;
@@ -500,7 +500,7 @@ SetLengthMods(tr,3);  // all equal
 			tr->langopts.param[LOPT_PREFIXES] = 1;
 			SetLetterVowel(tr,'y');
 		
-			tr->langopts.numbers = 0x11419;
+			tr->langopts.numbers = 0x11c19;
 			memcpy(tr->stress_lengths,stress_lengths_nl,sizeof(tr->stress_lengths));
 		}
 		break;
@@ -514,7 +514,7 @@ SetLengthMods(tr,3);  // all equal
 
 			tr->langopts.stress_rule = 0;
 			SetLetterVowel(tr,'y');
-			tr->langopts.numbers = 0x11049;
+			tr->langopts.numbers = 0x11849;
 		}
 		break;
 
@@ -531,7 +531,7 @@ SetLengthMods(tr,3);  // all equal
 			tr->langopts.stress_flags = 0x6;  // mark unstressed final syllables as diminished
 			tr->langopts.param[LOPT_REGRESSIVE_VOICING] = 0x8;
  			tr->langopts.max_initial_consonants = 7; // for example: wchrzczony :)
- 			tr->langopts.numbers=0x1809 + 0x4000;
+ 			tr->langopts.numbers=0x1009 + 0x4000;
 			tr->langopts.numbers2=0x40;
 			tr->langopts.param[LOPT_COMBINE_WORDS] = 2 + 0x100;  // combine 'nie' (marked with $alt2) with some 1-syllable words (marked with $alt)
 			SetLetterVowel(tr,'y');
@@ -549,7 +549,7 @@ SetLengthMods(tr,3);  // all equal
 //			tr->langopts.vowel_pause = 1;
 			tr->langopts.stress_rule = 3;        // stress on final syllable
 			tr->langopts.stress_flags =  0x6 | 0x10 | 0x20000; 
-			tr->langopts.numbers = 0xa69 + 0x2000 + NUM_ROMAN;
+			tr->langopts.numbers = 0x269 + 0x2000 + NUM_ROMAN;
 			SetLetterVowel(tr,'y');
 		}
 		break;
@@ -566,7 +566,7 @@ SetLengthMods(tr,3);  // all equal
 			tr->langopts.stress_flags = 0x100 + 0x6;
 
 			tr->charset_a0 = charsets[2];   // ISO-8859-2
-			tr->langopts.numbers = 0x1829+0x6000 + NUM_ROMAN;
+			tr->langopts.numbers = 0x1029+0x6000 + NUM_ROMAN;
 			tr->langopts.numbers2 = 0x1e;  // variant numbers before all thousandplex
 		}
 		break;
@@ -593,7 +593,7 @@ SetLengthMods(tr,3);  // all equal
 			tr->langopts.spelling_stress = 1;
 			tr->langopts.param[LOPT_COMBINE_WORDS] = 4;  // combine some prepositions with the following word
 
-			tr->langopts.numbers = 0x0c01 + 0x4000 + NUM_ROMAN;
+			tr->langopts.numbers = 0x0401 + 0x4000 + NUM_ROMAN;
 			tr->langopts.numbers2 = 0x40;
 			tr->langopts.thousands_sep = 0;   //no thousands separator
 			tr->langopts.decimal_sep = ',';
@@ -620,7 +620,7 @@ SetLengthMods(tr,3);  // all equal
 			tr->langopts.stress_rule = 0;
 			SetLetterVowel(tr,'y');
 //			SetLetterBits(tr,6,"eiyäö");   // soft vowels  NOTE accented letters don't work in SetLetterBits
-			tr->langopts.numbers = 0x1109;
+			tr->langopts.numbers = 0x1909;
 		}
 		break;
 
@@ -637,7 +637,7 @@ SetLengthMods(tr,3);  // all equal
 			tr->langopts.stress_rule = 2;
 			tr->langopts.stress_flags =  0x6 | 0x10; 
 
-			tr->langopts.numbers = 0x8e1;
+			tr->langopts.numbers = 0x4e1;
 			tr->langopts.numbers2 = 0x100;
 		}
 		break;
@@ -684,16 +684,17 @@ SetLengthMods(tr,3);  // all equal
 
 	case L('t','r'):   // Turkish
 		{
-			static const unsigned char stress_amps_tr[8] = {16,16, 20,20, 20,24, 24,22 };
-			static const short stress_lengths_tr[8] = {170,170, 190,170, 0,0, 250,270};
+			static const unsigned char stress_amps_tr[8] = {18,18, 20,20, 20,24, 24,22 };
+			static const short stress_lengths_tr[8] = {190,190, 190,190, 0,0, 250,270};
 
 			tr = new Translator();
 			SetupTranslator(tr,stress_lengths_tr,stress_amps_tr);
 			tr->charset_a0 = charsets[9];   // ISO-8859-9 - Latin5
 
 			tr->langopts.stress_rule = 7;   // stress on the last syllable, before any explicitly unstressed syllable
+			tr->langopts.stress_flags = 0x20;  //no automatic secondary stress
 
-			tr->langopts.numbers = 0x1d09 + 0x4000;
+			tr->langopts.numbers = 0x1509 + 0x4000;
 			tr->langopts.max_initial_consonants = 2;
 		}
 		break;
@@ -1044,7 +1045,7 @@ Translator_Afrikaans::Translator_Afrikaans() : Translator()
 	langopts.param[LOPT_PREFIXES] = 1;
 	SetLetterVowel(this,'y');  // add 'y' to vowels
 
-	langopts.numbers = 0x0d1 + NUM_ROMAN;
+	langopts.numbers = 0x8d1 + NUM_ROMAN;
 	memcpy(stress_lengths,stress_lengths2,sizeof(stress_lengths));
 }
 

src/translate.cpp

@@ -613,6 +613,8 @@ int Translator::TranslateWord(char *word1, int next_pause, WORD_TAB *wtab)
 	int word_length;
 	int ix;
 	int posn;
+	int pfix;
+	int n_chars;
 	unsigned int dictionary_flags[2];
 	unsigned int dictionary_flags2[2];
 	int end_type=0;
@@ -624,6 +626,7 @@ int Translator::TranslateWord(char *word1, int next_pause, WORD_TAB *wtab)
 	char prefix_phonemes[N_WORD_PHONEMES];
 	char end_phonemes[N_WORD_PHONEMES];
 	char word_copy[N_WORD_BYTES];
+	char prefix_chars[N_WORD_BYTES];
 	int found=0;
    int end_flags;
 	char c_temp;   // save a character byte while we temporarily replace it with space
@@ -824,7 +827,7 @@ if((wmark > 0) && (wmark < 8))
 			{
 				// Found a standard prefix, remove it and retranslate
 
-				if(confirm_prefix)
+				if(confirm_prefix && !(end_type & SUFX_B))
 				{
 					int end2;
 					char phonemes2[N_WORD_PHONEMES];
@@ -856,8 +859,6 @@ if((wmark > 0) && (wmark < 8))
 					}
 				}
 
-				strcat(prefix_phonemes,end_phonemes);
-				end_phonemes[0] = 0;
 				prefix_type = end_type;
 
 				if(prefix_type & SUFX_V)
@@ -866,15 +867,44 @@ if((wmark > 0) && (wmark < 8))
 				}
 
 				wordx[-1] = c_temp;
-				for(ix=(prefix_type & 0xf); ix>0; ix--)    // num. of characters to remove
+				pfix = 1;
+				prefix_chars[0] = 0;
+				n_chars = prefix_type & 0x3f;
+
+				for(ix=0; ix < n_chars; ix++)    // num. of characters to remove
 				{
-					wordx++;
-					while((*wordx & 0xc0) == 0x80) wordx++;  // for multibyte characters
+					prefix_chars[pfix++] = *wordx++;
+
+					if((prefix_type & SUFX_B) && (ix == (n_chars-1)))
+					{
+						prefix_chars[pfix-1] = 0;  // discard the last character of the prefix, this is the separator character
+					}
+
+					while((*wordx & 0xc0) == 0x80)
+					{
+						prefix_chars[pfix++] = *wordx++;  // for multibyte characters
+					}
 				}
+				prefix_chars[pfix] = 0;
 				c_temp = wordx[-1];
 				wordx[-1] = ' ';
 				confirm_prefix = 1;
 
+				if(prefix_type & SUFX_B)
+				{
+					// retranslate the prefix part
+					char *wordpf;
+					wordpf = &prefix_chars[1];
+					found = LookupDictList(&wordpf, phonemes, dictionary_flags, SUFX_P, wtab);   // without prefix
+					if(found == 0)
+					{
+						end_type = TranslateRules(wordpf, phonemes, N_WORD_PHONEMES, end_phonemes, 0, dictionary_flags[0]);
+						strcat(prefix_phonemes, phonemes);
+					}
+				}
+				strcat(prefix_phonemes,end_phonemes);
+				end_phonemes[0] = 0;
+
 				end_type = 0;
 				found = LookupDictList(&wordx, phonemes, dictionary_flags2, SUFX_P, wtab);   // without prefix
 				if(dictionary_flags[0]==0)

src/translate.h

@@ -106,6 +106,7 @@
 #define SUFX_F        0x2000   // verb follows
 #define SUFX_Q        0x4000   // don't retranslate
 #define SUFX_T        0x10000   // don't affect the stress position in the stem
+#define SUFX_B        0x20000  // break, this character breaks the word into stem and suffix (used with SUFX_P)
 
 #define FLAG_ALLOW_TEXTMODE  0x02  // allow dictionary to translate to text rather than phonemes
 #define FLAG_SUFX       0x04
@@ -321,7 +322,7 @@ typedef struct {
    // bit8=only one primary stress in tens+units
 	// bit9=only one vowel betwen tens and units
 	// bit10=omit "one" before "hundred"
-	// bit11=don't say 19** as nineteen hundred
+	// bit11=say 19** as nineteen hundred
 	// bit12=allow space as thousands separator (in addition to langopts.thousands_sep)
 	// bits13-15  post-decimal-digits 0=single digits, 1=(LANG=it) 2=(LANG=pl) 3=(LANG=ro)
 	// bit16=dot after number indicates ordinal