[1.31.19}

Fix: Events callback, fix word length information for words with multibyte characters.
Fix: Re-translating a word in a different language after the first langauge has started speaking individual letters, gave silence.
Announce punctuation.  If character name is not defined, use English.
Announce punctuation.  Fix pitch of the voice which announces punctuation characters.
[[Phonemes]] text input.  Allow the language-switch phoneme eg. [[_^_EN hEloU _^_FR]] in the input.


git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@152 d46cf337-b52f-0410-862d-fd96e6ae7743

dictsource/dict_phonemes

@@ -43,9 +43,9 @@ x    z
 Dictionary de_dict
 
 3    @    @-   a    A    A:   aI   aU   
-E    E2   E:   e:   EI   I    i2   i:   
-O    o:   OY   U    u:   W    y    y:   
-Y:   
+E    E2   E:   e:   EI   I    i2   I:   
+i:   O    o:   OY   U    u:   W    y    
+y:   Y:   
 
 *    :    ;    b    C    C2   d    D    
 dZ   f    g    h    j    k    l    m    
@@ -88,8 +88,8 @@ r    R    s    t    ts   v    z    Z
 
 Dictionary es_dict
 
-@-   a    aI   aU   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    m    n    
@@ -116,10 +116,10 @@ i    i:   o    O    O~   u    W    w^i
 W~   y    Y    
 
 *    :    ;    b    c    d    dZ   f    
-g    h    j    k    l    L    m    n    
-N    n2   n^   p    p2   r    R    s    
-S    t    t2   tS   v    w    z    Z    
-z2   z3   
+g    h    j    k    l    m    n    N    
+n2   n^   p    p2   r    R    s    S    
+t    t2   tS   v    w    z    Z    z2   
+z3   
 
 
 Dictionary hi_dict
@@ -281,7 +281,7 @@ u    U    uI   u~   y
 g    h    j    k    l    l^   m    n    
 N    n^   p    Q    r    R    r-   s    
 S    s#   s;   t    T    tS   ts   v    
-w    x    z    Z    
+w    w2   x    z    Z    
 
 
 Dictionary ro_dict
@@ -340,10 +340,10 @@ Dictionary zh_dict
 i    i.   i[   iou  N-   o    o-   ou   
 u    uai  y    
 
-(u)  :    f    h    j    k    kh   l    
-m    n    N    p    ph   r    s    s.   
-S;   t    th   ts   ts.  ts.  tS;  tS;  
-tsh  v    w    x    z    z.   
+:    f    j    k    kh   l    m    n    
+N    p    ph   r    R    s    s.   S;   
+t    th   ts   ts.  ts.  tS;  tS;  tsh  
+v    w    x    z    z.   
 
 
 Dictionary jbo_dict
@@ -417,7 +417,7 @@ Dictionary rw_dict
 
 a    e    i    o    u    
 
-*    b    B    c    d    dZ   f    g    
-h    j    J    k    l    m    n    N    
-n^   p    q    s    S    S;   t    tS   
-w    x    z    
+*    :    b    B    c    d    dZ   f    
+g    h    j    J    k    l    m    n    
+N    n^   p    q    s    S    S;   t    
+tS   v    w    x    z    

dictsource/en_list

@@ -57,6 +57,7 @@ _??	sImb@L
 _#9	tab
 _#32	speIs
 
+_ 	hA@dspeIs    // U+00A0  no-break-space
 _!	Ekskl@m'eIS@n
 _"	kwoUts
 _#	haS
@@ -65,8 +66,8 @@ _%	p3s'Ent
 ?5 _%	pVRs'Ent
 _&	amp@sand
 _'	praIm
-_(	ElbrakIt
-_)	A@brakIt
+_(	lEftbrakIt
+_)	raItbrakIt
 _*	ast@rIsk
 _+	plVs
 _,	k0m@
@@ -80,24 +81,72 @@ _=	i:kw@Lz
 _>	A@aNg@L
 _?	kwEstS@n
 _@	at|saIn
-_[	Elskwe@
+_[	lEftskwe@
 _\	bakstroUk
-_]	A@skwe@
+_]	raItskwe@
 _^	s3:kVmfl,Eks
 ?5 _^	sIRkVmfl,Eks
 __	'Vnd3sk,o@
 _`	bakpraIm
-_{	ElbreIs
+_{	lEftbreIs
 _|	vi:b'A@
-_}	A@breIs
+_}	raItbreIs
 _~	tIld@
-
+_¡	I2nv3:tI2dEkskl@m'eIS@n
+_¦	broUk@nb'A@
+_¨	daIa2r'i:sI2s
+_ª	fEmI2nI2n'O@dI2n@L
+_«	lEft'ang@LkwoUts
+_­	makr0n
+_²	su:p@skrIptt'u:
+_³	su:p@skrIptT'ri:
+_´	a2kju:taksEnt
+_·	mId@Ld0t
+_¸	sEd'Il@
+_¹	su:p@skrIptw'0n
+_º	maskjUlI2n'O@dI2n@L
+_»	raIt'ang@LkwoUts
+_¿	I2nv3:tI2dkw'EstS@n
+
+_‐	haIf@n
+_‑	haIf@n
+_‒	daS
+_–	daS
+_—	EmdaS
+_―	h0rIz,0nt@Lb'A@
+_‖	d,Vb@Lb'A@
+_‘	lEftkwoUt
+_’	raItkwoUt
+_‚	loUkwoUt
+_‛	lEftkwoUt
+_“	lEftkwoUts
+_”	raItkwoUts
+_„	loUkwoUts
+_‟	lEftkwoUts
+_†	dag3
+_‡	dVb@Ldag3
+_•	bUlI2t
+_‣	bUlI2t
+_․	d0t
+_‥	tu:d0ts
+_…	I2lIpsI2s
+_‰	p3:m'i:l
+_′	praIm
+_″	dVb@LpraIm
+_‴	trIp@LpraIm
+_‼	Ekskla2m'eIS@nz
+_₠	jU@roU
+
+¢	sEnts
 £	paUnd
 ¥	jEn
+₨	ru:pi:
+₩	w0n
 §	sEkS@n
-_¶	par@graf   // in en_rules
+¶	par@graf   // in en_rules
 ¤	kVr@nsI2saIn
 ©	k0pI2raIt
+®	rEdZIst3d
 °	dI2gri:z
 −	maIn@s
 ±	plVs@maIn@s
@@ -110,7 +159,14 @@ _¶	par@graf   // in en_rules
 €	jU@roUz
 ♯	SA@p
 ♭	flat
+¬	n0t_|saIn
 ə	SwA:
+æ	eI'i:
+ɛ	oUp@n'i:
+ɔ	oUp@n'oU
+ð	ED
+ʃ	ES
+ʒ	EZ
 
 // numeric
 
@@ -1112,6 +1168,7 @@ leant            lEnt
 leapt            lEpt
 lego             lEgoU
 leisure          l'EZ3
+lemme            lElI       // let me
 lemonade                       $3
 leo              li:oU
 leopard          lEp@d
@@ -2625,7 +2682,7 @@ although         OlD'oU        $pause
 and              %and          $pause $only
 as               ,az           $pause $only
 (as is)          ,az,Iz        $pause
-(as it is)       azIt'Iz_
+(as it is)       ,azIt,Iz      $u+
 because          bIk'Vz        $pause
 once                           $pause
 or               %O@           $pause $only
@@ -2717,7 +2774,7 @@ like             laIk          $strend2
 
 
 
-
+// ?6  'one' = [wVn], 'of' = [Vv]
 ?6 noone         noUwVn
 ?6 anyone        EnIwVn
 ?6 everyone      EvrIwVn

dictsource/en_rules

@@ -5130,7 +5130,7 @@
 	μ	m'ju:_
 	ν	n'ju:_
 	ξ	ks'aI_
-	ο	oUm'aIkr0n_
+	ο	'oUmIkr0n_
 
 .group 0xcf   // Greek letters
 
@@ -5146,7 +5146,7 @@
 	ψ	ps'aI_
 	ω	'oUmIg@_
 	ώ	'oUmIg@_
-	ό	oUm'aIkr0n_
+	ό	'oUmIkr0n_
 
 
 // default group for "non of the above"
@@ -5171,6 +5171,7 @@
 	ž	Z
 	þ	T
 	ð	D
+	ñ	n^
 
 
             \\           bakstroUk
@@ -5226,5 +5227,4 @@
         %%) % (_         _::
 
 // extra symbols (mostly in en_list)
-            ¶        par@graf
             ¶¶       par@grafs

dictsource/eo_list

@@ -1,6 +1,5 @@
 // This file is UTF-8 encoded
 
-
 // Numbers
 _0		nul
 _1		'unu

dictsource/es_list

@@ -15,27 +15,27 @@ _;	p,untoik'oma
 _:	d,Osp'untOs
 _!	TE*R'aRaDmi**aTj'On
 _?	TE*R'aRintE*R,oQaTj'On
-_¡	aB@-*'iRaDmi**aTj'On
-_¿	aB@-*'iRintE*R,oQaTj'On
+_¡	aB**'iRaDmi**aTj'On
+_¿	aB**'iRintE*R,oQaTj'On
 _=	iQw'al
 _<	men'ORke
 _>	maJ^'ORke
-_'	apOst@-*'ofo
+_'	apOst**'ofo
 _"	kom'iJ^as
 _+	s'iQnOdesum'aR
 _$	dolar
 _#	almoaD'iJ^a
 _*	aste*'isko
 _-	gJ^On
-__	suB@-*aJ^'aDo
+__	suB**aJ^'aDo
 _/	ba*Ra
 _\	b'a*RaimbERt'iDa
-_`	aT'EntOg@-*'aBe
-_(	,aB@-*epa**'Entesis
+_`	aT'EntOg**'aBe
+_(	,aB**epa**'Entesis
 _)	Tj,E*Rapa**'Entesis
-_[	,aB@-*ekORtS'ete
+_[	,aB**ekORtS'ete
 _]	Tj,E*RakORtS'ete
-_{	,aB@-*eJ^'aBe
+_{	,aB**eJ^'aBe
 _}	Tj,E*RaJ^'aBe
 _«	kom'iJ^as||iTkJ^'ERDas
 _»	kom'iJ^as||dE**'EtSas
@@ -47,7 +47,7 @@ _»	kom'iJ^as||dE**'EtSas
 @	a*R'oBa
 /	ba*Ra
 ©	kopi*R'aIt
-£	liB@-*as
+£	liB**as
 ¶	p'a*Rafo
 §	sEkTj'On
 
@@ -56,8 +56,8 @@ _»	kom'iJ^as||dE**'EtSas
 _0	Te*o
 _1	'uno
 _2	d'Os
-_3	t@-*'es
-_4	kw'at@-*o
+_3	t**'es
+_4	kw'at**o
 _5	T'inko
 _6	s'eIs
 _7	sj'ete
@@ -66,12 +66,12 @@ _9	nw'eve
 _1X	dj'ET
 _11	'OnTe
 _12	d'oTe
-_13	t@-*'eTe
+_13	t**'eTe
 _14	kat'ORTe
 _15	k'inTe
 _20	v'eInte
 _2X	v'eInt
-_3X	t@-*'eInta
+_3X	t**'eInta
 _4X	kwa*'Enta
 _5X	Tinkw'Enta
 _6X	sEs'Enta
@@ -339,8 +339,8 @@ _o	o
 o       o  $atend
 v	uve
 x	Ekis
-_y	,ig@-*'iEQa
-y	,ig@-*'iEQa  $atend
+_y	,ig**'iEQa
+y	,ig**'iEQa  $atend
 ch	tSe
 ll	Ejje
 á	'a||aTEntw'aDa

dictsource/es_rules

@@ -6,6 +6,7 @@
 // ?2  Latin America
 
 .L01 j w l r d g n m
+.L02 b d g n m         // change [s] to [z] before these
 
 .group a
        _) a (_       a
@@ -38,6 +39,7 @@
        _) d (_       de
           d          D
        _) d          d
+       l) d          d
        m) d          d
        n) d          d
 
@@ -142,7 +144,7 @@
           r          R
        _) r          R
        A) r (A       **
-       C) r (A       @-*
+       C) r (A       **
           rr         *R
 
 
@@ -150,6 +152,8 @@
        _) s (_       Ese
           s          s
        &) s (_S1     s
+          s (L02     z
+          s (gY      s
 
 
 .group t
@@ -199,7 +203,7 @@
 
 
 .group y
-       _) y (_       ,ig@-*'iEQa
+       _) y (_       ,ig**'iEQa
           y          i
        _) y          J
       _m) y          J

dictsource/fr_list

@@ -65,7 +65,7 @@ _y      i:gR'Ek
 
 //_cap	k,ap@-t@L
 _?A	lEt@
-_??	sE~b'OL
+_??	sE~b'Oll
 _#9	tabylasjO~
 _#32	Espas
 
@@ -229,9 +229,11 @@ prof                           $dot
 
 // articles
 le	$u $nounf
+le	$u $verb $verbextend
 la	$u $nounf
+la	$u $verb $verbextend
 les	lez2 $u $nounf
-?2 les	lez2 $u $nounf
+les	lez2 $u $verb $verbextend
 un	$u $nounf
 une	$u $nounf
 
@@ -278,7 +280,15 @@ lui	$u $verbf
 
 ce	$u $nounf
 cette	$u $nounf
-ces	sEz2 $u $nounf
+ces	sez2 $u $nounf
+ceci	$verbf
+cela	$verbf
+(celle ci)	sEl||si $verbf
+(celle là)	sEl||la $verbf
+(ceux ci)	sY||si $verbf
+(ceux là)	sY||la $verbf
+(celles ci)	sEl||si $verbf
+(celles là)	sEl||la $verbf
 
 // possessives
 ma	$u+ $nounf
@@ -292,16 +302,21 @@ son	$u+ $nounf
 ses	sez2 $u+ $nounf
 notre	$u+ $nounf
 votre	$u+ $nounf
-leur	$u+
+leur	$u+ $nounf
+leur	$u+ $verb $verbextend
+(la leur)	la||l@r $u+ $verbf
+(le leur)	l@||l@r $u+ $verbf
 nos	$u+ $nounf
 vos	$u+ $nounf
-leurs	$u+
+leurs	$u+ $nounf
+leurs	$u+ $verb $verbextend
+(les leur)	le||l@r $u+ $verbf
+(les leurs)	le||l@r $u+ $verbf
 
 // prepositions
 de $u $nounf
 du	$u+ $nounf
 des	dez2 $u $nounf
-?2 des  dez2 $u+ $nounf
 
 
 à	$u
@@ -321,7 +336,7 @@ avec	$u
 sans	$u
 sous	$u
 entre	$u $brk
-contre	$u $brk
+contre	$u
 
 durant	$brk
 
@@ -385,8 +400,7 @@ couvent          k'uvt2  $verb
 dessus          d@sy
 dessous          d@su
 divers          divErz2
-parent            paR'A~
-parent            p'art2  $verb
+hareng          _!arA~
 pastis     pastis
 poing     pwE~
 ressent          r@s'A~t2
@@ -438,6 +452,7 @@ sus       sy $verb
 vénus       venys
 
 
+gens	      ZA~
 mens	      mA~z2
 sens	      sA~s
 sens	      sA~z2 $verb
@@ -636,4 +651,3 @@ vincent		vE~sA~
 
 
 
-

dictsource/fr_rules

@@ -4,7 +4,7 @@
 
 // 2006-11-18  Gilles Casse  <[email protected]>
 //
-// Updated: 2008-02-24 Michel Such  <[email protected]>
+// Updated: 2008-02-26 Michel Such  <[email protected]>
 //
 //	* The rules are based on Cicero TTS.
 //         Y
@@ -140,6 +140,10 @@
 	_noeu)	d (_        		// noeud
 	r)	d (_        		// lourd placard
 		d           d		// don bled
+	_expé)	dien (t_    djA~		// expédient
+	V_expé)	dien (t_    di		// expédient
+	_ingré)	dien (t_    djA~		// ingrédient
+		dien (t_    di		// étudient (verbe)
 
 // group d: English section
 		day (_        _^_en
@@ -176,7 +180,6 @@
 	C)	en (d_      A~		// comprend dépend
 	od)	en (dr      E~		// rhododendron
 	C)	en (ds_     A~
-	cipi)	en (C     A~		// récipient
 	qu)	en (d_      A~
 	qu)	en (ds_     A~
 
@@ -194,7 +197,9 @@
 	_C)	en (t_      A~		// cent vent lent dent
 	cc)	en (t_      A~		// accent
 	Asc)	en (t_      A~		// luminescent
-	dol)	en (t_      A~		// dolent
+	_dol)	en (t_      A~		// dolent
+	ol)	en (t_      A~		// somnolent, violent
+	Vol)	en (t_        		// somnolent, violent
 	imCoC)	en (t_      A~		// impotent
 	inCoC)	en (t_      A~		// innocent, indolent
 
@@ -210,13 +215,21 @@
 	oémin)	en (t_      A~		// proéminent
 	jac)	en (t_      A~		// sous-jacent, adjacent
 	_lat)	en (t_      A~		// latent
-	lig)	en (t_      A~		// intelligent
+
+	@@Cig)	en (t_      A~		// intransigent, intelligent
+	_dilig)	en (t_      A~		// diligent
+	XACig)	en (t_      A~		// dirigent, exigent
+	VXACig)	en (t_        		// dirigent, exigent (verbe)
+	flig)	en (t_      A~		// affligent
+	Vflig)	en (t_        		// infligent (verbe)
+	Xig)	en (t_      		// figent, pigent
+	oblig)	en (t_      A~		// obligent, d‚sobligent
+	Voblig)	en (t_        		// obligent (verbe)
+
 	man)	en (t_      A~		// permanent
-	ndig)	en (t_      A~		// indigent
+	@@ing)	en (t_      A~		// astreingent, contingentindigent
 	_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
@@ -234,21 +247,26 @@
 	Atim)	en (t_      A~		// bâtiment
 	_cim)	en (t_      A~		// ciment
 	_pim)	en (t_      A~		// piment
+
+	ertin)	en (t_      A~		// pertinent, impertinent
+	ontin)	en (t_      A~		// continent, incontinent
 	erm)	en (t_      A~		// ferment, serment
 	Vferm)	en (t_      		// ferment, referment (verbe)
 	Arp)	en (t_      A~		// arpent, serpent
 	CArr)	en (t_      A~		// conncurrent, torrent
+	_appar)	en (t_      A~		// apparent
+	spar)	en (t_      A~		// transparent
+	_par)	en (t_      A~		// parent
+	V_par)	en (t_        		// parent (verbe)
+	_var)	en (t_      A~		// varent
 	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
-	oti)	en (t_      A~		// quotient
-
-	vi)	en (t_      E~
+	Cflu)	en (t_      A~		// affluent, confluent
+	VCflu)	en (t_        		// refluent, influent (verbe)
 
 	al)	en (t_      A~		// talent équivalent
+	Vival)	en (t_        		// équivalent (verbe)
 	xcell)	en (t_      A~		// excellent
 	nn)	en (t_      t2		// prennent, viennent, sonnent
 
@@ -281,7 +299,7 @@
 		en (t_            	// étaient mangent
 
 	Ci)	en (_       E~      	// viendra
-	Ci)	en (C       E~      	// tien
+	Ci)	en (C       E~      	// viendra
 	Ci)	en (c       A~      	// science
 
 
@@ -435,11 +453,11 @@
 		ed (_  	_^_en
 	_)	edit   	_^_en
 		ee          _^_en		// meeting
-		eing       _^_en		// being, goeing
+		eing (_      _^_en		// being, boeing
 		eive       _^_en		// receive
 		ej       _^_en
 	 	eld (_        _^_en		// field
-	 	enter (_        _^_en		// enter, center
+	_) 	enter (_        _^_en		// enter
 	_) 	enhan         _^_en
 	ack)	et (_      _^_en     	// racket, packet
 	ock)	et (_      _^_en     	// rocket, pocket
@@ -605,6 +623,15 @@ _)		half  	_^_en
 		i           i		// cri
 
 		ie (ment    i		// remerciement balbutiement
+	C)	ien (t_       i
+	dév)	ien (t_       i
+	t)	ien (t_      jE~		// vient
+	v)	ien (t_      jE~		// tient
+	at)	ien (t_      jA~		// patient
+	cip)	ien (C     jA~		// récipient
+	vén)	ien (C     jA~		// inconvénient
+	ot)	ien (t_      jA~		// quotient
+
 		i (es_      i		// parties
 		i (e_       i
 		i (A        j		// fermier portier patio renier
@@ -688,7 +715,7 @@ _)		half  	_^_en
 		mm          m		// pomme
 
 // group m: English section
-		mail (A       _^_en  // mailer
+	_)	mail (_       _^_en  // mailer
 		mov (A         _^_en		// move, movy
 
 
@@ -971,9 +998,9 @@ C)		te (_      t@-  		// patte
 
 	a)	t (ia       s   // spatial
 	en)	t (ia       s   // confidentialit�
-	u)	t (ia       s   // balbutiait
+	bu)	t (iA       s   // balbutiait
 	r)	t (ia       s   // impartial
-	i)	t (ia       s   // initia
+	ini)	t (i       s   // initier
 	p)	t (ia       s   // nuptial
 	an)	t (iat      s   // transsubstantiation
 
@@ -993,8 +1020,9 @@ C)		te (_      t@-  		// patte
  	pa) 	t (ien      s
  	o) 	t (ien      s
 
-	Cu) 	t (ie_      s	// minutie
-	Cu) 	t (ies_     s
+	Cu) 	t (ieX      s	// minutie
+	ni) 	tien (t_     si
+	bu) 	tien (t_     si
 
 	cra) 	t (ie_      s	// démocratie
 	cra) 	t (ies_     s

dictsource/grc_rules

@@ -8,6 +8,7 @@
 //   Replace Latin characters by same-shape Greek characters
 
 .replace
+′	'
 a	α  // Latin->Greek
 o	ο
 v	ν
@@ -26,6 +27,32 @@ T	Τ
 Y	Υ
 X	Χ
 
+// replace Latin characters by Greek equivalents
+th	θ
+ph	φ
+kh	χ
+ps	ψ
+b	β
+g	γ
+d	δ
+e	ε
+z	ζ
+q	η // (long e; not standard transliteration but may be easier to type than η)
+i	ι
+k	κ
+l	λ
+m	μ
+n	ν
+x	ξ
+p	π
+r	ρ
+s	σ
+t	τ
+u	υ
+y	υ
+w	ω // (long o; not standard transliteration but may be easier to type
+
+
 // alpha
 ά	ὰ    // tonos
 ά	ὰ    // oxia
@@ -161,184 +188,175 @@ X	Χ
 
 
 .group α
-        α        a
-        αι       aI
-        αὶ      'aI
-        αυ       aU
-        αὺ      'aU
-        αhι      haI
-        αhὶ     'haI
-        αhυ      haU
-        αhὺ     'haU
+	α        a
+	αι       aI
+	αὶ      'aI
+	αυ       aU
+	αὺ      'aU
+	αhι      haI
+	αhὶ     'haI
+	αhυ      haU
+	αhὺ     'haU
 
 .group β
-        β        b
+	β        b
 
 
 .group γ
-        γ        g
-        γγ       Ng
+	γ        g
+	γγ       Ng
 
 .group δ
-        δ        d
+	δ        d
 
 
 .group ε
-        ε        e
-        ει       e:
-        εὶ      'e:
-        ευ       eU
-        εὺ      'eU
-        εhι      he:
-        εhὶ     'he:
-        εhυ      heU
-        εhὺ     'heU
+	ε        e
+	ει       e:
+	εὶ      'e:
+	ευ       eU
+	εὺ      'eU
+	εhι      he:
+	εhὶ     'he:
+	εhυ      heU
+	εhὺ     'heU
 
 .group ζ
-        ζ        z
+	ζ        z
 
 
 .group η
-        η        E:
-        ηι       EI:
-        ηὶ      'EI:
-        ηυ       EU:
-        ηὺ      'EU:
-        ηhι      hEI:
-        ηhὶ     'hEI:
-        ηhυ      hEU:
-        ηhὺ     'hEU:
+	η        E:
+	ηι       EI:
+	ηὶ      'EI:
+	ηυ       EU:
+	ηὺ      'EU:
+	ηhι      hEI:
+	ηhὶ     'hEI:
+	ηhυ      hEU:
+	ηhὺ     'hEU:
 
 .group θ
-        θ        T
+	θ        T
 
 
 .group ι
-        ι        i
+	ι        i
 
 
 .group ϊ
-        ϊ        i
+	ϊ        i
 
 
 .group κ
-        κ        k
+	κ        k
 
 
 .group λ
-        λ        l
+	λ        l
 
 
 .group μ
-        μ        m
+	μ        m
 
 
 .group ν
-        ν        n
+	ν        n
 
 
 .group ξ
-        ξ        ks
+	ξ        ks
 
 
 .group ο
-        ο        o
-        οι       oI
-        οὶ      'oI
-        ου       u:
-        οὺ      'u:
-        οhι      hoI
-        οhὶ     'hoI
-        οhυ      hu:
-        οhὺ     'hu:
+	ο        o
+	οι       oI
+	οὶ      'oI
+	ου       u:
+	οὺ      'u:
+	οhι      hoI
+	οhὶ     'hoI
+	οhυ      hu:
+	οhὺ     'hu:
 
 .group π
-        π        p
+	π        p
 
 
 .group ρ
-        ρ        R
+	ρ        R
 
 
 .group ς
-        ς        s
+	ς        s
 
 
 .group σ
-        σ        s
+	σ        s
 
 
 .group τ
-        τ        t
+	τ        t
 
 
 .group υ
-        υ        y
-        υι       yI
-        υὶ      'yI
-        υhι      hyI
-        υhὶ     'hyI
+	υ        y
+	υι       yI
+	υὶ      'yI
+	υhι      hyI
+	υhὶ     'hyI
 
 .group ϋ
-        ϋ        y
-        ϋι       yI
-        ϋὶ      'yI
-        ϋhι      hyI
-        ϋhὶ     'hyI
+	ϋ        y
+	ϋι       yI
+	ϋὶ      'yI
+	ϋhι      hyI
+	ϋhὶ     'hyI
 
 .group φ
-        φ        f
+	φ        f
 
 
 .group χ
-        χ        x
+	χ        x
 
 
 .group ψ
-        ψ        ps
+	ψ        ps
 
 
 .group ω
-        ω        O:
-        ωι       OI:
-        ωὶ      'OI:
-        ωυ       OU:
-        ωὺ      'OU:
-        ωhι      hOI:
-        ωhὶ     'hOI:
-        ωhυ      hOU:
-        ωhὺ     'hOU:
+	ω        O:
+	ωι       OI:
+	ωὶ      'OI:
+	ωυ       OU:
+	ωὺ      'OU:
+	ωhι      hOI:
+	ωhὶ     'hOI:
+	ωhυ      hOU:
+	ωhὺ     'hOU:
 
 
 
 .group h   // dasia (rough breathing)
-        h        h
+	h        h
 
 
 .group
-        ὰ       'a
-        ᾳ        aI:
-        ᾲ       'aI:
-        ὲ       'e
-        ὴ       'E:
-        ῃ        EI:
-        ῂ       'EI:
-        ὶ       'i
-        ῒ       'i
-        ὸ       'o
-        ὺ       'y
-        ῢ       'y
-        ὼ       'O:
-        ῳ        OI:
-        ῲ       'OI:
-
-
-
-
-
-
-
-
-
-
-
+	ὰ       'a
+	ᾳ        aI:
+	ᾲ       'aI:
+	ὲ       'e
+	ὴ       'E:
+	ῃ        EI:
+	ῂ       'EI:
+	ὶ       'i
+	ῒ       'i
+	ὸ       'o
+	ὺ       'y
+	ῢ       'y
+	ὼ       'O:
+	ῳ        OI:
+	ῲ       'OI:
+
+	'	=	// apostrophe used as tonos, stress on previous syllable

dictsource/pt_list

@@ -34,8 +34,8 @@ x	Sis#
 
 
 _cap	maI'uskulU
-_?A	l'etR&
-_??	s'imbolU
+_?A	l'etR&_
+_??	s'imbolU_
 
 _.	p'oNtU
 _!	,eskl&mas'&U~

dictsource/ta_rules

@@ -55,7 +55,7 @@
 	க (் 	g
      ்) க	ga
      ்) க (B	g
-     _) க	kV
+     _) க	ka
      _) க (B	k
 	க்க	k:a
 	க்க (B	k:
@@ -78,7 +78,7 @@
 
 	ட	d.a
 	ட (B	d.
-     _) ட	t.V
+     _) ட	t.a
      _) ட (B	t.
 	ட்ட	t.a
 	ட்ட (B	t.
@@ -88,7 +88,7 @@
 
 	த	da
 	த (B	d
-     _) த	tV
+     _) த	ta
      _) த (B	t
 	த்த	t:a
 	த்த (B	t:
@@ -101,7 +101,7 @@
 
 	ப	ba
 	ப (B	b
-     _) ப	pV
+     _) ப	pa
      _) ப (B	p
 	ப்ப	p:a
 	ப்ப (B	p:

dictsource/zh_rules

@@ -82,7 +82,7 @@ language).
 
 .group c
        c        tsh
-       ch       ts.h
+       ch (+    ts.h
     @) c (K	_^_EN
 
 .group d
@@ -150,6 +150,7 @@ language).
 .group l
        l        l
     @) l (K	_^_EN
+    @) l (v     l
 
 .group m
        m        m
@@ -188,7 +189,7 @@ language).
 
 .group s
        s        s
-       sh       s.
+       sh (+    s.
     @) s (K	_^_EN
 
 .group t
@@ -239,7 +240,7 @@ language).
     n) ve	yE //üe
 
 .group w
-//    @) w (K	_^_EN
+    @) w (K	_^_EN
        wa	wA  //wa wan wang
        wai	wai
        wa1i	wai55
@@ -270,7 +271,7 @@ language).
 
 .group z
        z        ts
-       zh       ts.
+       zh (+    ts.
     @) z (K	_^_EN
 
 //tone number

phsource/compile_report

@@ -1,55 +1,55 @@
 50 phoneme tables
           new total
-    base  99   99
-   base2  24  118
-      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
-      eo  13  111
-     jbo   4  112
-      fi  40  130
-      fr  44  125
-   fr_ca  11  125
-      hi  51  136
-      ta  16  139
-      hu  23  115
-      nl  28  122
-      pl  15  110
-      sk  29  128
-      cs   6  128
-      hr  20  133
-      mk   3  134
-      sr  14  133
-      ru  38  129
-      it  17  121
-      la  21  117
-      es   9  119
-   es_la   1  119
-      pt  27  134
-   pt_pt  20  134
-      ro  36  142
-      el   8  118
-     grc  12  123
-      sv  25  121
-      no  28  125
-      is  32  124
-      vi  42  137
-     zhy  40  132
-      zh  65  149
-      sw  14  108
-      th  50  142
-      id  14  120
-      tr  18  123
-      ku  13  120
-      ja   7  104
-      rw  12  123
+    base  98   98
+   base2  24  117
+      en  51  144
+    en_n  30  144
+   en_us  34  144
+   en_sc  39  146
+   en_rp  34  144
+   en_wm  30  144
+   en_wi  30  144
+      af  38  130
+      cy  29  124
+      de  33  125
+      eo  13  110
+     jbo   4  111
+      fi  40  129
+      fr  44  124
+   fr_ca  11  124
+      hi  51  135
+      ta  16  138
+      hu  23  114
+      nl  28  121
+      pl  15  109
+      sk  29  127
+      cs   6  127
+      hr  20  132
+      mk   3  133
+      sr  14  132
+      ru  38  128
+      it  17  120
+      la  21  116
+      es   9  118
+   es_la   1  118
+      pt  27  133
+   pt_pt  20  133
+      ro  36  141
+      el   8  117
+     grc  12  122
+      sv  25  120
+      no  28  124
+      is  32  123
+      vi  42  136
+     zhy  40  131
+      zh  65  148
+      sw  14  107
+      th  50  141
+      id  14  119
+      tr  18  122
+      ku  13  119
+      ja   7  103
+      rw  12  122
 
 Data file      Used by
 b/b             [b] base
@@ -501,6 +501,7 @@ r3/rx           [*] base
 r3/rz_cs        [R^] sk
                 [R^/] sk
 r3/@tap         [**] base
+r3/@tap2        [**] base
 r/r             [r] base
                 [r-] base
                 [r] en_sc
@@ -1024,7 +1025,6 @@ vdiph/@u_3      [@U] ro
 vdiph/@u_en     [oU] en
                 [@U] id
 vdiph/ui        [uI] base2
-                [uI] en
                 [uI] eo
                 [ui] fi
                 [uI] vi
@@ -1083,8 +1083,7 @@ vnasal/oo_n2    [O~] hi
 vnasal/u_n      [u~] hi
                 [u~] pt
                 [u] zh
-vnasal/W_n      [W~] en
-                [E~] fr
+vnasal/W_n      [E~] fr
                 [W~] fr
 voc/bh          [B] base
 voc/dh          [D] base
@@ -1455,8 +1454,7 @@ vowel/i_5       [i:] en_sc
 vowel/i#_5      [y] pt_pt
                 [i/] pt_pt
                 [y] ro
-vowel/i_6       [i:] en_us
-                [i] ta
+vowel/i_6       [i] ta
                 [i] hr
                 [i] id
                 [i] tr
@@ -1464,6 +1462,7 @@ vowel/i_6       [i:] en_us
 vowel/i#_6      [i.] zh
 vowel/i#_7      [i[] zh
 vowel/i_en      [i:] en
+                [i:] en_us
                 [i:] fr
                 [i] vi
 vowel/i_fnt     [i:] en_wi

phsource/ph_english

@@ -292,7 +292,7 @@ endphoneme
 
 phoneme e@
   vowel  starttype (e)  endtype (@)
-  length 220
+  length 230
   formants vowel/ee_1
   linkout r-
 endphoneme
@@ -361,13 +361,6 @@ formants vdiph/eu
 endphoneme
 
 
-phoneme uI
-  vowel  starttype (u)  endtype (i)
-  length 240
-  formants vdiph/ui
-endphoneme
-
-
 phoneme IR        // used for "ir" when [3:] is split (Scottish)
   vowel  starttype (@)  endtype (@)
   length 190
@@ -405,13 +398,6 @@ phoneme O~
 endphoneme
 
 
-phoneme W~
-  vowel  starttype (@)  endtype (@)
-  length 260
-  formants vnasal/W_n
-endphoneme
-
-
 phoneme y:
   vowel  starttype (u)  endtype (u)
   length 220

phsource/ph_english_us

@@ -136,7 +136,7 @@ endphoneme
 phoneme i:
   vowel  starttype (i)  endtype (i)
   length 170
-  formants vowel/i_6
+  formants vowel/i_en
   linkout ;
 endphoneme
 

phsource/phonemes

@@ -31,7 +31,6 @@ phonemenumber 24 _::    // long pause
 phonemenumber 25 t#     // reduced version of [t]
 phonemenumber 26 '!     // stress - emphasized
 phonemenumber 27 _;_    // clause pause
-phonemenumber 28 _^^    // Change language + phoneme table number
 
 // The base phoneme table contains the utility phonemes, the consonants,
 // and a schwa.  The vowels are defined in the language specific phoneme
@@ -150,13 +149,6 @@ phoneme  _^_  //  Change dictionary (followed by language-code in ascii
   lengthmod 1
 endphoneme 
 
-phoneme  _^^  //  Change phoneme table (followed by phoneme table number)
-  pause
-  starttype _ endtype _
-  length 1
-  lengthmod 1
-endphoneme 
-
 phoneme  _X1  //  a language specific action
   pause
   starttype _ endtype _
@@ -605,9 +597,7 @@ phoneme **    //   TEST flap
   vowelout f1=3 f2=1600 -300 300  f3=-300 80 rms=35 len=15
   vowelin f1=2  f2=1600 -300 300  f3=-100 80 len=20
   formants r3/@tap
-//  vowelin f1=3  f2=1600 -300 300  f3=-300 80
-//  before (i) r3/@tap_i+r3/rx
-//  before (e) r3/@tap_i+r3/rx
+  after (e) r3/@tap2
 endphoneme
 
 

src/dictionary.cpp

@@ -434,11 +434,18 @@ char *EncodePhonemes(char *p, char *outptr, unsigned char *bad_phoneme)
 				while(!isspace(c = *p++) && (c != 0))
 					*outptr++ = tolower(c);
 				*outptr = 0;
-				if(strcmp(p_lang,"en")==0)
+				if(c == 0)
 				{
-					*p_lang = 0;   // don't need "en", it's assumed by default
+					if(strcmp(p_lang,"en")==0)
+					{
+						*p_lang = 0;   // don't need "en", it's assumed by default
+						return(p);
+					}
+				}
+				else
+				{
+					*outptr++ = '|';  // more phonemes follow, terminate language string with separator
 				}
-				return(p);
 			}
 			break;
 		}
@@ -2302,10 +2309,10 @@ int Translator::TranslateRules(char *p_start, char *phonemes, int ph_size, char
 	
 	while(((c = *p) != ' ') && (c != 0))
 	{
-		if(IsAlpha(wc))
-			any_alpha++;
 		wc_prev = wc;
 		wc_bytes = utf8_in(&wc,p,0);
+		if(IsAlpha(wc))
+			any_alpha++;
 
 		n = groups2_count[c];
 		if(IsDigit(wc) && ((langopts.tone_numbers == 0) || !any_alpha))
@@ -2424,11 +2431,23 @@ int Translator::TranslateRules(char *p_start, char *phonemes, int ph_size, char
 				if(match1.points == 0)
 				{
 	static const char str_unknown[4] = {phonCAPITAL,phonCAPITAL,phonCAPITAL,0};
-					if(((any_alpha > 0) || (p[wc_bytes-1] > ' ')) && !iswpunct(wc))
+					if(iswalpha(wc))
 					{
-						// an unrecognised character in a word, indicate with clicks
-						match1.phonemes = str_unknown;
-						match1.points = 1;
+						if((any_alpha > 1) || (p[wc_bytes-1] > ' '))
+						{
+							// an unrecognised character in a word, indicate with clicks
+							match1.phonemes = str_unknown;
+							match1.points = 1;
+						}
+					}
+					else
+					{
+						LookupLetter(wc, -1, ph_buf);
+						if(ph_buf[0])
+						{
+							match1.phonemes = ph_buf;
+							match1.points = 1;
+						}
 					}
 					p += (wc_bytes-1);
 				}

src/numbers.cpp

@@ -45,6 +45,31 @@ void Translator::LookupLetter(unsigned int letter, int next_byte, char *ph_buf1)
 	char ph_stress[2];
 	char ph_buf3[30];
 
+	ph_buf1[0] = 0;
+	len = utf8_out(letter,&single_letter[2]);
+	single_letter[len+2] = ' ';
+
+	if(next_byte == -1)
+	{
+		// speaking normal text, not individual characters
+		if(Lookup(&single_letter[2],ph_buf1) != 0)
+			return;
+
+		single_letter[1] = '_';
+		if(Lookup(&single_letter[1],ph_buf3) != 0)
+			return;   // the character is specified as _* so ignore it when speaking normal text
+
+		// check whether this character is specified for English
+		SetTranslator2("en");
+		if(translator2->Lookup(&single_letter[2], ph_buf3) != 0)
+		{
+			// yes, switch to English and re-translate the word
+			sprintf(ph_buf1,"%c",phonSWITCH);
+		}
+		SelectPhonemeTable(voice->phoneme_tab_ix);  // revert to original phoneme table
+		return;
+	}
+
 	if((letter <= 32) || iswspace(letter))
 	{
 		// lookup space as _&32 etc.
@@ -53,10 +78,6 @@ void Translator::LookupLetter(unsigned int letter, int next_byte, char *ph_buf1)
 		return;
 	}
 
-
-	len = utf8_out(letter,&single_letter[2]);
-	single_letter[len+2] = ' ';
-
 	if(next_byte != ' ')
 		next_byte = RULE_SPELLING;
 	single_letter[3+len] = next_byte;   // follow by space-space if the end of the word, or space-0x31
@@ -91,8 +112,8 @@ void Translator::LookupLetter(unsigned int letter, int next_byte, char *ph_buf1)
 
 
 
-int Translator::TranslateLetter(char *word, char *phonemes, int control)
-{//=====================================================================
+int Translator::TranslateLetter(char *word, char *phonemes, int control, int word_length)
+{//======================================================================================
 // get pronunciation for an isolated letter
 // return number of bytes used by the letter
 // control 2=say-as glyphs, 3-say-as chars
@@ -134,11 +155,17 @@ int Translator::TranslateLetter(char *word, char *phonemes, int control)
 		return(0);
 	}
 
-	if(ph_buf[0] == 0)
+	if((ph_buf[0] == 0) && (word_length == 1) && (translator_name != L('e','n')))
 	{
-// ?? speak as English ??
-		sprintf(phonemes,"%c",phonSWITCH);
-		return(0);
+		// speak as English, check whether there is a translation for this character
+		SetTranslator2("en");
+		translator2->LookupLetter(letter, word[n_bytes], ph_buf);
+		SelectPhonemeTable(voice->phoneme_tab_ix);  // revert to original phoneme table
+		if(ph_buf[0] != 0)
+		{
+			sprintf(phonemes,"%c",phonSWITCH);
+			return(0);
+		}
 	}
 
 	if(ph_buf[0] == 0)

src/readclause.cpp

@@ -63,11 +63,11 @@ static const char *punct_stop = ".:!?";    // pitch fall if followed by space
 static const char *punct_close = ")]}>;'\"";  // always pitch fall unless followed by alnum
 
 // alter tone for announce punctuation or capitals
-static const char *tone_punct_on = "\001+50R\001+10T";  // add reverberation, reduce low frequencies
-static const char *tone_punct_off = "\001R\001T";
+static const char *tone_punct_on = "\0016T";  // add reverberation, lower pitch
+static const char *tone_punct_off = "\001T";
 
 // punctuations symbols that can end a clause
-static const unsigned short punct_chars[] = {',','.','?','!',':',';',
+const unsigned short punct_chars[] = {',','.','?','!',':',';',
   0x2013,  // en-dash
   0x2014,  // em-dash
   0x2026,  // elipsis
@@ -477,22 +477,38 @@ static void UngetC(int c)
 }
 
 
+const char *WordToString2(unsigned int word)
+{//========================================
+// Convert a language mnemonic word into a string
+	int  ix;
+	static char buf[5];
+	char *p;
+
+	p = buf;
+	for(ix=3; ix>=0; ix--)
+	{
+		if((*p = word >> (ix*8)) != 0)
+			p++;
+	}
+	*p = 0;
+	return(buf);
+}
 
-const char *Translator::LookupSpecial(const char *string)
-{//======================================================
+
+const char *Translator::LookupSpecial(const char *string, char* text_out)
+{//======================================================================
 	unsigned int flags[2];
 	char phonemes[55];
 	char phonemes2[55];
-	static char buf[60];
 	char *string1 = (char *)string;
 
 	if(LookupDictList(&string1,phonemes,flags,0,NULL))
 	{
 		SetWordStress(phonemes,flags[0],-1,0);
 		DecodePhonemes(phonemes,phonemes2);
-		sprintf(buf,"[[%s]] ",phonemes2);
+		sprintf(text_out,"[[%s]]",phonemes2);
 		option_phoneme_input = 1;
-		return(buf);
+		return(text_out);
 	}
 	return(NULL);
 }
@@ -503,26 +519,78 @@ const char *Translator::LookupCharName(int c)
 // Find the phoneme string (in ascii) to speak the name of character c
 
 	int ix;
-	const char *p;
-	static char buf[24];
+	unsigned int flags[2];
+	char single_letter[24];
+	char phonemes[60];
+	char phonemes2[60];
+	char *lang_name = NULL;
+	char *string;
+	static char buf[60];
 
-	buf[0] = '_';
-	ix = utf8_out(c,&buf[1]);
-	buf[1+ix]=0;
+	buf[0] = 0;
+	flags[0] = 0;
+	flags[1] = 0;
+	single_letter[0] = 0;
+	single_letter[1] = '_';
+	ix = utf8_out(c,&single_letter[2]);
+	single_letter[2+ix]=0;
 
-	if((p = LookupSpecial(buf)) == NULL)
+	string = &single_letter[1];
+	if(LookupDictList(&string, phonemes, flags, 0, NULL) == 0)
 	{
-		p = LookupSpecial(&buf[1]);
+		// try _* then *
+		string = &single_letter[2];
+		if(LookupDictList(&string, phonemes, flags, 0, NULL) == 0)
+		{
+			// now try the rules
+			single_letter[1] = ' ';
+			TranslateRules(&single_letter[2], phonemes, sizeof(phonemes), NULL,0,0);
+		}
 	}
-	if(p != NULL)
-		return(p);
 
-	if((p = LookupSpecial("_??")) == NULL)
+	if((phonemes[0] == 0) && (translator_name != L('e','n')))
 	{
-		p = "symbol";
+		// not found, try English
+		SetTranslator2("en");
+		string = &single_letter[1];
+		single_letter[1] = '_';
+		if(translator2->LookupDictList(&string, phonemes, flags, 0, NULL) == 0)
+		{
+			string = &single_letter[2];
+			translator2->LookupDictList(&string, phonemes, flags, 0, NULL);
+		}
+		if(phonemes[0])
+		{
+			lang_name = "en";
+		}
+		else
+		{
+			SelectPhonemeTable(voice->phoneme_tab_ix);  // revert to original phoneme table
+		}
 	}
-	strcpy(buf,p);
-//	sprintf(buf,"%s%d ",p,c);
+
+	if(phonemes[0])
+	{
+		if(lang_name)
+		{
+			translator2->SetWordStress(phonemes,flags[0],-1,0);
+			DecodePhonemes(phonemes,phonemes2);
+			sprintf(buf,"[[_^_%s %s _^_%s]]","en",phonemes2,WordToString2(translator_name));
+			SelectPhonemeTable(voice->phoneme_tab_ix);  // revert to original phoneme table
+		}
+		else
+		{
+			SetWordStress(phonemes,flags[0],-1,0);
+			DecodePhonemes(phonemes,phonemes2);
+			sprintf(buf,"[[%s]] ",phonemes2);
+		}
+	}
+	else
+	{
+		strcpy(buf,"[[(X1)(X1)(X1)]]");
+	}
+
+	option_phoneme_input = 1;
 	return(buf);
 }
 
@@ -714,7 +782,9 @@ int Translator::AnnouncePunctuation(int c1, int c2, char *buf, int bufix)
 
 			p = &buf[bufix];
 			if(punct_count==1)
+			{
 				sprintf(p,"%s %s %s",tone_punct_on,punctname,tone_punct_off);
+			}
 			else
 			if(punct_count < 4)
 			{
@@ -1769,8 +1839,8 @@ int Translator::ReadClause(FILE *f_in, char *buf, unsigned short *charix, int n_
 	const char *p;
 	wchar_t xml_buf[N_XML_BUF+1];
 
-#define N_XML_BUF2   12
-	char buf2[N_XML_BUF2+2];
+#define N_XML_BUF2   20
+	char xml_buf2[N_XML_BUF2+2];           // for &<name> and &<number> sequences
 	static char ungot_string[N_XML_BUF2+4];
 	static int ungot_string_ix = -1;
 
@@ -1859,26 +1929,26 @@ f_input = f_in;  // for GetC etc
 				c1 = c2;
 				while(!Eof() && (iswalnum(c1) || (c1=='#')) && (n_xml_buf < N_XML_BUF2))
 				{
-					buf2[n_xml_buf++] = c1;
+					xml_buf2[n_xml_buf++] = c1;
 					c1 = GetC();
 				}
-				buf2[n_xml_buf] = 0;
+				xml_buf2[n_xml_buf] = 0;
 				c2 = GetC();
-				sprintf(ungot_string,"%s%c%c",&buf2[0],c1,c2);
+				sprintf(ungot_string,"%s%c%c",&xml_buf2[0],c1,c2);
 
 				if(c1 == ';')
 				{
-					if(buf2[0] == '#')
+					if(xml_buf2[0] == '#')
 					{
 						// character code number
-						if(buf2[1] == 'x')
-							found = sscanf(&buf2[2],"%x",(unsigned int *)(&c1));
+						if(xml_buf2[1] == 'x')
+							found = sscanf(&xml_buf2[2],"%x",(unsigned int *)(&c1));
 						else
-							found = sscanf(&buf2[1],"%d",&c1);
+							found = sscanf(&xml_buf2[1],"%d",&c1);
 					}
 					else
 					{
-						if((found = LookupMnem(xml_char_mnemonics,buf2)) != -1)
+						if((found = LookupMnem(xml_char_mnemonics,xml_buf2)) != -1)
 						{
 							c1 = found;
 							if(c2 == 0)
@@ -2050,14 +2120,15 @@ f_input = f_in;  // for GetC etc
 			clause_upper_count++;
 			if((option_capitals == 2) && !iswupper(cprev))
 			{
-				p = LookupSpecial("_cap");
-				if(p != NULL)
+				char text_buf[40];
+				char text_buf2[30];
+				if(LookupSpecial("_cap",text_buf2) != NULL)
 				{
-					sprintf(buf2,"%s%s%s",tone_punct_on,p,tone_punct_off);
-					j = strlen(buf2);
+					sprintf(text_buf,"%s%s%s",tone_punct_on,text_buf2,tone_punct_off);
+					j = strlen(text_buf);
 					if((ix + j) < n_buf)
 					{
-						strcpy(&buf[ix],buf2);
+						strcpy(&buf[ix],text_buf);
 						ix += j;
 					}
 				}
@@ -2186,11 +2257,14 @@ if(option_ssml) parag=1;
 		if(speech_parameters[espeakSILENCE]==1)
 			continue;
 
+		j = ix+1;
+		ix += utf8_out(c1,&buf[ix]);    //	buf[ix++] = c1;
 		if(!iswspace(c1) && !IsBracket(c1))
 		{
 			charix[ix] = count_characters - clause_start_char;
+			while(j < ix)
+				charix[j++] = 0xffff;   // subsequent bytes of a multibyte character
 		}
-		ix += utf8_out(c1,&buf[ix]);    //	buf[ix++] = c1;
 
 		if(((ix > (n_buf-20)) && !IsAlpha(c1) && !iswdigit(c1))  ||  (ix >= (n_buf-2)))
 		{

src/synthdata.cpp

@@ -35,7 +35,7 @@
 #include "translate.h"
 #include "wave.h"
 
-const char *version_string = "1.31.18  25.Feb.08";
+const char *version_string = "1.31.19  28.Feb.08";
 const int version_phdata  = 0x013105;
 
 int option_device_number = -1;

src/synthesize.h

@@ -51,7 +51,7 @@
 #define EMBED_A     3   // amplitude/volume
 #define EMBED_R     4   // pitch range/expression
 #define EMBED_H     5   // echo/reverberation
-#define EMBED_T     6   // different tone
+#define EMBED_T     6   // different tone for announcing punctuation
 #define EMBED_I     7   // sound icon
 #define EMBED_S2    8   // speed (used in synthesize)
 #define EMBED_Y     9   // say-as commands
@@ -274,5 +274,6 @@ int DoSample(PHONEME_TAB *ph1, PHONEME_TAB *ph2, int which, int length_mod, int
 int DoSpect(PHONEME_TAB *this_ph, PHONEME_TAB *prev_ph, PHONEME_TAB *next_ph,
 		int which, PHONEME_LIST *plist, int modulation);
 int PauseLength(int pause);
+int LookupPhonemeTable(const char *name);
 
 void InitBreath(void);

src/tr_english.cpp

@@ -111,7 +111,7 @@ int Translator_English::Unpronouncable(char *word)
 		if(c == '\'')
 			apostrophe = 1;
 		else
-		if((c < 'a') || (c > 0x241))
+		if(!IsAlpha(c))
 			return(0);        // letter (not vowel) outside Latin character range or apostrophe, abort test
 	}
 	if((vowel_posn > 5) || ((word[0]!='s') && (vowel_posn > 4)))

src/translate.cpp

@@ -752,7 +752,7 @@ if((wmark > 0) && (wmark < 8))
 
 		while(*wordx != ' ')
 		{
-			wordx += TranslateLetter(wordx, phonemes,spell_word);
+			wordx += TranslateLetter(wordx, phonemes,spell_word, word_length);
 			posn++;
 			if(phonemes[0] == phonSWITCH)
 			{
@@ -779,7 +779,7 @@ if((wmark > 0) && (wmark < 8))
 			// This word looks "unpronouncable", so speak letters individually until we
 			// find a remainder that we can pronounce.
 			emphasize_allcaps = 0;
-			wordx += TranslateLetter(wordx,phonemes,0);
+			wordx += TranslateLetter(wordx,phonemes,0, word_length);
 			posn++;
 			if(phonemes[0] == phonSWITCH)
 			{
@@ -825,12 +825,12 @@ if((wmark > 0) && (wmark < 8))
 				// characters not recognised, speak them individually
 
 				utf8_in(&wc, wordx, 0);
-				if(!iswpunct(wc))
+				if((word_length == 1) && IsAlpha(wc))
 				{
 					posn = 0;
 					while(*wordx != ' ')
 					{
-						wordx += TranslateLetter(wordx, phonemes, 4);
+						wordx += TranslateLetter(wordx, phonemes, 4, word_length);
 						posn++;
 						if(phonemes[0] == phonSWITCH)
 						{
@@ -1278,9 +1278,12 @@ int Translator::TranslateWord2(char *word, WORD_TAB *wtab, int pre_pause, int ne
 	int first_phoneme = 1;
 	int source_ix;
 	int len;
+	int ix;
 	int sylimit;        // max. number of syllables in a word to be combined with a preceding preposition
 	const char *new_language;
 	unsigned char bad_phoneme[4];
+	int word_copy_len;
+	char word_copy[N_WORD_BYTES];
 
 	len = wtab->length;
 	if(len > 31) len = 31;
@@ -1355,11 +1358,42 @@ int Translator::TranslateWord2(char *word, WORD_TAB *wtab, int pre_pause, int ne
 	if(word_flags & FLAG_PHONEMES)
 	{
 		// The input is in phoneme mnemonics, not language text
-		EncodePhonemes(word,word_phonemes,bad_phoneme);
+		int c1;
+		char lang_name[12];
+
+		if(memcmp(word,"_^_",3)==0)
+		{
+			// switch languages
+			word+=3;
+			for(ix=0;;)
+			{
+				c1 = *word++;
+				if((c1==' ') || (c1==0))
+					break;
+				lang_name[ix++] = tolower(c1);
+			}
+			lang_name[ix] = 0;
+
+			if((ix = LookupPhonemeTable(lang_name)) > 0)
+			{
+				SelectPhonemeTable(ix);
+				word_phonemes[0] = phonSWITCH;
+				word_phonemes[1] = ix;
+				word_phonemes[2] = 0;
+			}
+		}
+		else
+		{
+			EncodePhonemes(word,word_phonemes,bad_phoneme);
+		}
 		flags = FLAG_FOUND;
 	}
 	else
 	{
+		ix = 0;
+		while((word_copy[ix] = word[ix]) != ' ') ix++;
+		word_copy_len = ix;
+
 		flags = translator->TranslateWord(word, next_pause, wtab);
 
 		if((flags & FLAG_ALT2_TRANS) && ((sylimit = langopts.param[LOPT_COMBINE_WORDS]) > 0))
@@ -1420,6 +1454,8 @@ int Translator::TranslateWord2(char *word, WORD_TAB *wtab, int pre_pause, int ne
 		if(p[0] == phonSWITCH)
 		{
 			// this word uses a different language
+			memcpy(word, word_copy, word_copy_len);
+
 			new_language = (char *)(&p[1]);
 			if(new_language[0]==0)
 				new_language = "en";
@@ -1523,6 +1559,15 @@ int Translator::TranslateWord2(char *word, WORD_TAB *wtab, int pre_pause, int ne
 
 		// Add the phonemes to the first stage phoneme list (ph_list2)
 		ph = phoneme_tab[ph_code];
+
+		if(ph_code == phonSWITCH)
+		{
+			ph_list2[n_ph_list2].phcode = ph_code;
+			ph_list2[n_ph_list2].sourceix = 0;
+			ph_list2[n_ph_list2].synthflags = embedded_flag;
+			ph_list2[n_ph_list2++].tone_number = *p++;
+		}
+		else
 		if(ph->type == phSTRESS)
 		{
 			// don't add stress phonemes codes to the list, but give their stress
@@ -1924,8 +1969,14 @@ void *Translator::TranslateClause(FILE *f_text, const void *vp_input, int *tone_
 
 	for(j=0; charix[j]==0; j++);
 	words[0].sourceix = charix[j];
-	for(k=j; charix[k]!=0; k++);
-	words[0].length = k-j;
+	k = 0;
+	while(charix[j] != 0)
+	{
+		// count the number of characters (excluding multibyte continuation bytes)
+		if(charix[j++] != 0xffff)
+			k++;
+	}
+	words[0].length = k;
 
 	while(!finished && (ix < (int)sizeof(sbuf))&& (n_ph_list2 < N_PHONEME_LIST-4))
 	{
@@ -2327,10 +2378,16 @@ if((c == '/') && (langopts.testing & 2) && IsDigit09(next_in) && IsAlpha(prev_ou
 				words[word_count].start = ix;
 				words[word_count].flags = 0;
 
-				for(j=source_index; charix[j] == 0; j++);
+				for(j=source_index; charix[j] == 0; j++);   // skip blanks
 				words[word_count].sourceix = charix[j];
-				for(k=j; charix[k]!=0; k++);
-				words[word_count].length = k-j;
+				k = 0;
+				while(charix[j] != 0)
+				{
+					// count the number of characters (excluding multibyte continuation bytes)
+					if(charix[j++] != 0xffff)
+						k++;
+				}
+				words[word_count].length = k;
 
 				word_flags = next_word_flags;
 				next_word_flags = 0;

src/translate.h

@@ -417,7 +417,7 @@ public:
 
 private:
 	int TranslateWord2(char *word, WORD_TAB *wtab, int pre_pause, int next_pause);
-	int TranslateLetter(char *letter, char *phonemes, int control);
+	int TranslateLetter(char *letter, char *phonemes, int control, int word_length);
 	void SetSpellingStress(char *phonemes, int control, int n_chars);
 	void WriteMnemonic(int *ix, int mnem);
 	void MakePhonemeList(int post_pause, int new_sentence);
@@ -427,7 +427,7 @@ private:
 	int AnnouncePunctuation(int c1, int c2, char *buf, int ix);
 
 	const char *LookupDict2(const char *word, const char *word2, char *phonetic, unsigned int *flags, int end_flags, WORD_TAB *wtab);
-	const char *LookupSpecial(const char *string);
+	const char *LookupSpecial(const char *string, char *text_out);
 	const char *LookupCharName(int c);
 	void LookupLetter(unsigned int letter, int next_byte, char *ph_buf);
 	int LookupNum2(int value, int control, char *ph_out);
@@ -539,6 +539,7 @@ extern char skip_marker[N_MARKER_LENGTH];
 #define N_PUNCTLIST  60
 extern wchar_t option_punctlist[N_PUNCTLIST];  // which punctuation characters to announce
 extern unsigned char punctuation_to_tone[INTONATION_TYPES][PUNCT_INTONATIONS];
+extern const unsigned short punct_chars[];   // punctuation chars fo end-of-clause
 
 extern int speech_parameters[];
 

src/wavegen.cpp

@@ -144,7 +144,7 @@ int wcmdq_head=0;
 int wcmdq_tail=0;
 
 // pitch,speed,
-int embedded_default[N_EMBEDDED_VALUES]        = {0,50,170,100,50, 0,50, 0,170,0,0,0,0,0};
+int embedded_default[N_EMBEDDED_VALUES]        = {0,50,170,100,50, 0,0, 0,170,0,0,0,0,0};
 static int embedded_max[N_EMBEDDED_VALUES]     = {0,0x7fff,360,300,99,99,99, 0,360,0,0,0,0,4};
 
 #define N_CALLBACK_IX N_WAV_BUF-2   // adjust this delay to match display with the currently spoken word
@@ -813,6 +813,12 @@ static void WavegenSetEcho(void)
 		amp = embedded_value[EMBED_H];
 		delay = 130;
 	}
+	if(embedded_value[EMBED_T] > 0)
+	{
+		// announcing punctuation
+		amp = embedded_value[EMBED_T] * 10;
+		delay = 130;
+	}
 
 	if(delay == 0)
 		amp = 0;
@@ -1517,8 +1523,9 @@ void SetEmbedded(int control, int value)
 
 	switch(command)
 	{
-	case EMBED_P:
 	case EMBED_T:
+		WavegenSetEcho();   // and drop through to case P
+	case EMBED_P:
 		// adjust formants to give better results for a different voice pitch
 		if((pitch_value = embedded_value[EMBED_P]) > MAX_PITCH_VALUE)
 			pitch_value = MAX_PITCH_VALUE;
@@ -1528,7 +1535,7 @@ void SetEmbedded(int control, int value)
 		{
 			wvoice->freq[ix] = (wvoice->freq2[ix] * factor)/256;
 		}
-		factor = (embedded_value[EMBED_T] - 50)*2;
+		factor = embedded_value[EMBED_T]*3;
 		wvoice->height[0] = (wvoice->height2[0] * (256 - factor*2))/256;
 		wvoice->height[1] = (wvoice->height2[1] * (256 - factor))/256;
 		break;
@@ -1623,6 +1630,7 @@ if(option_log_frames)
 
 	if((pitch_value = embedded_value[EMBED_P]) > MAX_PITCH_VALUE)
 		pitch_value = MAX_PITCH_VALUE;
+	pitch_value -= embedded_value[EMBED_T];   // adjust tone for announcing punctuation
 	if(pitch_value < 0)
 		pitch_value = 0;