[1.29.18] Add data for Serbian voice ("sr").

Moved Croatian/Bosnian/Serbian dictionary source data to hbs_rules, hbs_list.
Added data for Ancient Greek voice ("grc").
In voice files, the attribute: word n  adds a short pause between words. n = 1 to 4 for increasing pause length.


git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@103 d46cf337-b52f-0410-862d-fd96e6ae7743

dictsource/dict_phonemes

 th   v    w    x    z    
 
 
-Dictionary hr_dict
-
-&    @    @-   @2   a    A    a:   aI   
-aU   E    e    e:   E~   i    I    i:   
-l-   O    o    o:   oU   r-   u    U    
-u:   y    
-
-*    b    d    dZ   dZ;  f    g    h    
-j    k    l    l^   m    n    N    n^   
-p    r    R    s    S    t    tS   ts   
-tS;  v    x    z    Z    
-
-
 Dictionary hu_dict
 
-A    a:   E    e:   i    i:   o    o:   
-u    u:   Y    Y:   
-
-:    b    c    d    dZ   f    g    h    
-j    J    k    l    m    n    n^   p    
-R    R2   s    S    s2   t    tS   ts   
-v    z    Z    
 
 
 Dictionary it_dict
 p    ph   R    R2   R3   s    s.   S;   
 t    th   ts   ts.  ts.  tS;  tS;  tsh  
 v    w    x    z    z.   
+
+
+Dictionary hbs_dict
+
+&    @    @2   a    A    a:   aI   aU   
+E    e    e:   E~   i    I    i:   l-   
+O    o    o:   oU   r*   r-   u    U    
+u:   y    
+
+*    b    d    dZ   dZ;  f    g    h    
+j    k    l    L    l^   m    n    N    
+n^   p    r    R    s    S    t    tS   
+ts   tS;  v    x    z    Z    

dictsource/en_list

 
 naif             naI'i:f
 naïve            naI'i:v
+nasa             nasa2
 nascent          neIs@nt
 nativity                       $2
 natural          natS@r@L

dictsource/en_rules

             as (tern     a2s
             as (thm      as
             as (tral     as
-            as (tro      aas
+            as (tro      as
             as (trolog   a2s
             as (trono    a2s
             ass (ym      as

dictsource/grc_list

+
+?1  _0lang   _^_EN   // speak numbers in English
+
+// articles
+hο	$u	// ὁ
+τὼ	$u	// τώ
+οhι	$u	// οἱ
+hη	$u	// ἡ
+τὰ	$u	// τά
+αhι	$u	// αἱ
+τὸ	$u	// τό
+
+τοὺ	$u	// τοῦ
+τοὶν	$u	// τοῖν
+τὴς	$u	// τῆς	
+ταὶν	$u	// ταῖν
+τὼν	$u	// τῶν
+
+τῲ	$u	// τῷ
+τοὶς	$u	// τοῖς
+τῂ	$u	// τῇ
+ταὶν	$u	// ταῖν
+ταὶς	$u	// ταῖς
+
+τὸν	$u	// τόν
+τοὺς	$u	// τούς
+τὴν	$u	// τήν
+τὰς	$u	// τάς
+

dictsource/grc_rules

+// Spelling to Phoneme rules for: Ancient Greek
+
+// Preliminary character substitution:
+//   Convert tonos, oxia and perispomeni to varia
+//   Remove psili (soft breathing)
+//   Convert dasia accent to a separate prefix character ("h")
+//   Remove iota subscripts
+//   Replace Latin characters by same-shape Greek characters
+
+.replace
+a	α  // Latin->Greek
+o	ο
+v	ν
+A	Α
+B	Β
+E	Ε
+Z	Ζ
+H	Η
+I	Ι
+K	Κ
+M	Μ
+N	Ν
+O	Ο
+P	Ρ
+T	Τ
+Y	Υ
+X	Χ
+
+// alpha
+ά	ὰ    // tonos
+ά	ὰ    // oxia
+ᾴ	ᾲ
+ᾶ	ὰ
+ᾷ	ᾲ
+
+ἀ	α
+ἁ	hα
+ἂ	ὰ
+ἃ	hὰ
+ἄ	ὰ
+ἅ	hὰ
+ἆ	ὰ
+ἇ	hὰ
+
+ᾀ	ᾳ
+ᾁ	hᾳ
+ᾂ	ᾲ
+ᾃ	hᾲ
+ᾄ	ᾲ
+ᾅ	hᾲ
+ᾆ	ᾲ
+ᾇ	hᾲ
+
+// epsilon
+έ	ὲ
+έ	ὲ
+ἐ	ε
+ἑ	hε
+ἒ	ὲ
+ἓ	hὲ
+ἔ	ὲ
+ἕ	hὲ
+
+// eta
+ή	ὴ
+ή	ὴ
+ῄ	ῂ
+ῆ	ὴ
+ῇ	ῂ
+
+ἠ	η
+ἡ	hη
+ἢ	ὴ
+ἣ	hὴ
+ἤ	ὴ
+ἥ	hὴ
+ἦ	ὴ
+ἧ	hὴ
+
+ᾐ	ῃ
+ᾑ	hῃ
+ᾒ	ῂ
+ᾓ	hῂ
+ᾔ	ῂ
+ᾕ	hῂ
+ᾖ	ῂ
+ᾗ	hῂ
+
+//iota
+ί	ὶ
+ί	ὶ
+ῖ	ὶ
+
+ἰ	ι
+ἱ	hι
+ἲ	ὶ
+ἳ	hὶ
+ἴ	ὶ
+ἵ	hὶ
+ἶ	ὶ
+ἷ	hὶ
+
+ΐ	ῒ
+ῗ	ῒ
+
+// omicron
+ό	ὸ
+ό	ὸ
+ὀ	ο
+ὁ	hο
+ὂ	ὸ
+ὃ	hὸ
+ὄ	ὸ
+ὅ	hὸ
+
+// upsilon
+ύ	ὺ
+ύ	ὺ
+ῦ	ὺ
+
+ὐ	υ
+ὑ	hυ
+ὒ	ὺ
+ὓ	hὺ
+ὔ	ὺ
+ὕ	hὺ
+ὖ	ὺ
+ὗ	hὺ
+
+ΰ	ῢ
+ῧ	ῢ
+
+// omega
+ώ	ὼ
+ώ	ὼ
+ῴ	ῲ
+ῶ	ὼ
+ῷ	ῲ
+
+ὠ	ω
+ὡ	hω
+ὢ	ὼ
+ὣ	hὼ
+ὤ	ὼ
+ὥ	hὼ
+ὦ	ὼ
+ὧ	hὼ
+
+ᾠ	ῳ
+ᾡ	hῳ
+ᾢ	ῲ
+ᾣ	hῲ
+ᾤ	ῲ
+ᾥ	hῲ
+ᾦ	ῲ
+ᾧ	hῲ
+
+// rho
+ῤ	ρ
+ῥ	hρ
+
+
+.group α
+        α        a
+        αι       aI
+        αὶ      'aI
+        αυ       aU
+        αὺ      'aU
+        αhι      haI
+        αhὶ     'haI
+        αhυ      haU
+        αhὺ     'haU
+
+.group β
+        β        b
+
+
+.group γ
+        γ        g
+        γγ       Ng
+
+.group δ
+        δ        d
+
+
+.group ε
+        ε        e
+        ει       e:
+        εὶ      'e:
+        ευ       eU
+        εὺ      'eU
+        εhι      he:
+        εhὶ     'he:
+        εhυ      heU
+        εhὺ     'heU
+
+.group ζ
+        ζ        z
+
+
+.group η
+        η        E:
+        ηι       EI:
+        ηὶ      'EI:
+        ηυ       EU:
+        ηὺ      'EU:
+        ηhι      hEI:
+        ηhὶ     'hEI:
+        ηhυ      hEU:
+        ηhὺ     'hEU:
+
+.group θ
+        θ        T
+
+
+.group ι
+        ι        i
+
+
+.group ϊ
+        ϊ        i
+
+
+.group κ
+        κ        k
+
+
+.group λ
+        λ        l
+
+
+.group μ
+        μ        m
+
+
+.group ν
+        ν        n
+
+
+.group ξ
+        ξ        ks
+
+
+.group ο
+        ο        o
+        οι       oI
+        οὶ      'oI
+        ου       u:
+        οὺ      'u:
+        οhι      hoI
+        οhὶ     'hoI
+        οhυ      hu:
+        οhὺ     'hu:
+
+.group π
+        π        p
+
+
+.group ρ
+        ρ        R
+
+
+.group ς
+        ς        s
+
+
+.group σ
+        σ        s
+
+
+.group τ
+        τ        t
+
+
+.group υ
+        υ        y
+        υι       yI
+        υὶ      'yI
+        υhι      hyI
+        υhὶ     'hyI
+
+.group ϋ
+        ϋ        y
+        ϋι       yI
+        ϋὶ      'yI
+        ϋhι      hyI
+        ϋhὶ     'hyI
+
+.group φ
+        φ        f
+
+
+.group χ
+        χ        x
+
+
+.group ψ
+        ψ        ps
+
+
+.group ω
+        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
+
+
+.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:
+
+
+
+
+
+
+
+
+
+
+

dictsource/hbs_list

+
+// "hbs" is the ISO 639-3 code for Serbian/Croatian/Bosnian macrolanguage
+// This file is UTF-8 encoded
+
+// Conditional rules:
+// ?1  Croatian
+// ?2  Serbian
+// ?3  Bosnian (Bosna and Hercegovina)
+// ?4  Serbian & Bosnian
+
+
+// Letters
+   b	b@
+   c	ts@
+   č	tS@
+   ć	tS;@2
+   d	d@
+   dž	dZ@
+   đ	dZ;@2
+   f	f@
+   g	g@
+   h	x@
+   j	j@2
+   k	k@
+   l	l@
+   lj	l^@
+   m	m@
+   n	n@
+   nj	n^@
+   p	p@
+   q	kv@
+   r	R@
+   _s	s@
+   š	S@
+   t	t@
+   v	v@
+   w	dvostr*uko||v@
+?2 w	duplo||v@
+   x	iks
+   y	ipsilon
+   z	z@
+   ž	Z@
+
+// symbols
+   _??	znak          // unknown symbol
+   _?A	slovo         // unknown letter
+   ©	'aUtoRsk&||pr*av&
+   %	p'osto
+   +	plus
+   €	'euRo
+?4 €	'evr*o
+   @	_'at
+?2 @	tRgovatSkoi
+   &	_'end
+   !	'usklItSnIk
+?2 !	'uzvItSnIk
+   :	dv'ototSk&
+?4 :	dv'ot&tSk&
+   #	br*'oj_
+   /	kr*'oz_
+   \\	b'EkslES
+   ^	ts'iRkumfleks
+   ~	t'ild&
+   ¤	v'alut&
+   §	'odlom&k
+   µ	m'ikr*o
+   ¶	tSl'an&k
+   °	st'up&n^
+?4 °	st'epen
+   _"	n'avodnIk
+?2 _"	n'avod
+   _‚	'otvoReni||n'avodnIk
+   _„	sp'uSteni||n'avodnIk
+?2 _„	z&tvoReni||n'avodnIk
+   …	tr*'i||t'otSke
+?4 …	tr*'i||t'atSke
+   †	kr*'iZ
+?2 †	kr-'st
+   ‡	dv'ostr*Uki||kr*'iZ
+?2 ‡	dv'ostRUki||kR'st
+   ‰	pr*'omil
+   _‹	'otvoReni||n'avodnIk
+   _‘	'otvoReni||n'avodnIk
+   _’	j'ednostr*'uki||n'avodnIk
+   _“	'otvoReni||n'avodnIk
+   _”	z'atvoReni||n'avodnIk
+   •	kr*'upna||t'otSk&
+?4 •	kr*'upna||t'atSk&
+   _–	En||ts'r-t&
+   _—	Em||ts'r-t&
+   ™	pr*'o'izvodZ&tS
+   ˘	br*'evis
+   ¨	pr*'ijegl&s
+?2 ¨	pr*'egl&s
+   _«	'otvoReni||n'avodnIk
+   _­	pr*'ivr*emen&||ts'r-t'its&
+   ·	p'ut&
+   ¸	sed'il&
+   _»	z'atvoReni||n'avodnIk
+   ®	R'egistr*atsIj&
+   ł	l@||s&||ts'r-t'itsom
+   ×	p'ut&
+   ÷	p'odijel^,eno
+?2 ÷	p'odel^,eno
+
+   _$    d'ol&R
+   _-	m'inus
+   _*	zvj'ezdits&
+?2 _*	zv'ezdits&
+   _=	j'edn&ko
+   _@	t'r-g'ov&tSko||'a
+?2 _@	et
+
+   _(	'otvoRen&||z'agr*ad&
+   _)	z'atvoRen&||z'agr*ad&
+   _'	'apostr*of
+   _,	z'aRez
+?2 _,	z'apeta
+   _-	ts'r-t'its&
+   _.	t'otSk&
+?4 _.	t'atSk&
+   _;	t'otSk&z,a*Ez
+?2 _;	t'atSk&z,apeta
+?3 _;	t'atSk&z,a*Ez
+   _<	m'an^i||'od
+   _=	j'edn&ko
+   _>	v'etSi||'od
+   _?	'upitnIk
+   _&	t'r-g'ov&tSko||'i
+   _[	'otvoRen&||'ugl&t&
+?2 _[	'otvoRen&||'ugl&st&
+   _]	z'atvoRen&||'ugl&t&
+?2 _]	z'atvoRen&||'ugl&st&
+   __	ts'r-t&
+   _`	gr*'avis
+   _{	'otvoRen&||v'ititS&st&
+   _|	'okomits&
+?2 _|	'uspR&vn&
+   _}	z'atvoRen&||v'ititS&st&
+
+
+// Numbers
+   _0    n'ul&
+   _1    j'ed&n
+   _2    dv'a
+   _2f   dv'ije
+?2 _2f   dv'E
+   _3    tr*'i
+   _4    tS'EtIRI
+   _5    p'Et
+   _6    S'Est
+   _7    s'Ed&m
+   _8    'os&m
+   _9    d'EvEt
+   _10   d'EsEt
+   _11   j'ed&naIst
+   _12   dv'anaIst
+   _13   tr*'inaIst
+   _14   tS'Etr-naIst
+   _15   p'EtnaIst
+   _16   S'EsnaIst
+   _17   s'Ed&mnaIst
+   _18   'os&mnaIst
+   _19   d'EvEtnaIst
+?2 _11   j'ed&naEst
+?2 _12   dv'anaEst
+?2 _13   tr*'inaEst
+?2 _14   tS'Etr-naEst
+?2 _15   p'EtnaEst
+?2 _16   S'EsnaEst
+?2 _17   s'Ed&mnaEst
+?2 _18   'os&mnaEst
+?2 _19   d'EvEtnaEst
+   _2X   dv'adEsEt
+   _3X   tr*'idEsEt
+   _4X   tS'Etr-dEsEt
+   _5X   p'EdEsEt
+   _6X   S'EzdEsEt
+   _7X   s'Ed&mdEsEt
+   _8X   'os&mdEsEt
+   _9X   d'EvEdEsEt
+   _0C   st'o_
+?2 _0C   st'O_
+   _2C   dvj'est'o
+?2 _2C   dv'est'a_
+?2 _3C   tR'ist'a_
+?2 _4C   tSetR'ist'o_
+   _6C   S'Est'o
+   _0M1  t'isUtS;&
+   _0MA1 t'isUtS;e
+   _1M1  t'isUtS;U
+?2 _0M1  h'il^,&d&
+?2 _0MA1 h'il^,&de
+?2 _1M1  h'il^,&dU
+?2 _1MA1  jedn&||hiL&d&
+?3 _0M1  h'il^ad&
+?3 _0MA1 h'il^ade
+?3 _1M1  h'il^adU
+   _0M2  m'ilijUna
+   _0MA2 m'ilijUna
+   _1M2  m'ilijUn
+?4 _0M2  m'ilijOna
+?4 _0MA2 m'ilijOna
+?4 _1M2  m'ilijOn
+   _0M3  m'iliaRdI
+   _0MA3 m'iliaRde
+   _1M3  m'iliaRdU
+?2 _0M3  milijArdI
+?2 _0MA3 milijArde
+?2 _1M3  milijArdU
+   _0M4  b'ilijUna
+   _1M4  b'ilijUn
+?4 _0M4  b'ilijOna
+?4 _1M4  b'ilijOn
+   _dpt  _:z'a*Ez_
+   _roman  r*'imsko
+
+// not Roman numerals
+// vi	vi   // not needed if only all-capitals are spoken as Roman numbers
+
+
+// Abbreviations
+aaa	$abbrev
+aac	$abbrev
+ab	$only $abbrev $capital
+abc	$abbrev
+abk	$only $abbrev
+abs	$abbrev
+acc $only $abbrev
+acf $only $abbrev
+acfm $abbrev
+acg $only $abbrev
+acgm $abbrev
+acl $only $abbrev
+acm $only $abbrev
+acw $only $abbrev
+add	$only $abbrev
+adn $only $abbrev
+adp $only $abbrev
+adsl	$abbrev
+agm	$abbrev
+ahk $only $abbrev
+aif $only $abbrev
+aip $only $abbrev
+alg $only $abbrev
+alm $only $abbrev
+amd	$only $abbrev
+amf $only $abbrev
+amg	$only $abbrev
+aph $only $abbrev
+apj $only $abbrev
+apl $only $abbrev
+apm $only $abbrev
+app $only $abbrev
+arg	$only $abbrev
+arj	$only $abbrev
+asc $only $abbrev
+asf $only $abbrev
+asm $only $abbrev
+asn $only $abbrev
+asp $only $abbrev
+asx $only $abbrev
+atf $only $abbrev
+atp	$abbrev
+av	$only $abbrev
+avg	$abbrev
+avr $only $abbrev
+awb $only $abbrev
+awt $only $abbrev
+brf	$abbrev
+brl	$abbrev
+brx	$abbrev
+cca	$abbrev
+cda $only $abbrev
+cdr	$abbrev
+cdrw	$abbrev
+cgi $abbrev
+clr	$abbrev
+cpe $only $abbrev
+cpu $only $abbrev
+cr	$only $abbrev
+crt $only $abbrev
+ctrl	$abbrev
+daa $only $abbrev
+doo	$abbrev
+dox $only $abbrev
+dqy $abbrev
+dsr $only $abbrev
+ear $only $abbrev
+ecdl	$only $abbrev
+ecp $only $abbrev
+ecs $only $abbrev
+elm $only $abbrev
+emf $only $abbrev
+emh $only $abbrev
+eml $only $abbrev
+ems $only $abbrev
+epf $only $abbrev
+eq $only $abbrev
+esc	$only $abbrev
+etd $only $abbrev
+eu	$only $abbrev
+eur	$only $abbrev
+exc $only $abbrev
+faq	$abbrev
+fbi	$abbrev
+ffa $only $abbrev
+fr	$only $abbrev
+frm	$abbrev
+frt $only $abbrev
+frx	$abbrev
+gho $only $abbrev
+gr	$only $abbrev
+grr	$abbrev
+grrr	$abbrev
+grp	$abbrev
+grps	$abbrev
+grs $only $abbrev
+grx $only $abbrev
+gui $only $abbrev
+hho	$abbrev
+hr	$only $abbrev
+hrk	$only $abbrev
+hrt	$only $abbrev $capital
+hrv	$only $abbrev
+hta $only $abbrev
+hzzo	$abbrev
+ibm	$abbrev
+icc $only $abbrev
+icm $only $abbrev
+ics $only $abbrev
+idb $only $abbrev
+idc $only $abbrev
+idm $only $abbrev
+iff $only $abbrev
+ifu $only $abbrev
+ilk $only $abbrev
+img $only $abbrev
+iml $only $abbrev
+isdn $abbrev
+isl $only $abbrev
+isp $only $abbrev
+iss $only $abbrev
+iq $only $abbrev
+iqy $only $abbrev
+itw $only $abbrev
+itz $only $abbrev
+ivf $only $abbrev
+jpi $only $abbrev
+jrr	$abbrev
+jse $only $abbrev
+lha $abbrev
+liq $only $abbrev
+maq $only $abbrev
+maw $only $abbrev
+mda $only $abbrev
+mde $only $abbrev
+mdi $only $abbrev
+mga $only $abbrev
+mge $only $abbrev
+mgi $only $abbrev
+mpa $only $abbrev
+mpe $only $abbrev
+mpr $only $abbrev
+msi $only $abbrev
+mup $abbrev $capital
+nco $only $abbrev
+nfo $only $abbrev
+nji $only $abbrev
+nr	$only $abbrev
+nra $only $abbrev
+nrb $only $abbrev
+nrc $only $abbrev
+nre $only $abbrev
+nrf $only $abbrev
+nrg	$abbrev
+nrh $only $abbrev
+nri $only $abbrev
+nrm $only $abbrev
+nrq $only $abbrev
+nrs $only $abbrev
+nru $only $abbrev
+nrw $only $abbrev
+nsa $only $abbrev
+nsi $only $abbrev
+nvda	$abbrev
+obd $only $abbrev
+obj $only $abbrev
+obt $only $abbrev
+obz $only $abbrev
+ocr $only $abbrev
+ocx $only $abbrev
+oem	$abbrev
+off	$only $abbrev
+oft $only $abbrev
+ofx $only $abbrev
+ogc $only $abbrev
+ogg	$only $abbrev
+ogl $only $abbrev
+omw	$abbrev
+oqy $only $abbrev
+os $only $abbrev $capital
+oss $only $abbrev
+otf $only $abbrev
+otm $only $abbrev
+pda $abbrev
+pdi $only $abbrev
+pgi $only $abbrev
+pho $only $abbrev
+pko $only $abbrev
+pma $only $abbrev
+ppa $only $abbrev
+pr	$only $abbrev
+prf $only $abbrev
+prg $only $abbrev
+prn	$abbrev
+prt $only $abbrev
+qbr $abbrev
+qpr $abbrev
+raw	$only $abbrev
+rb	$abbrev
+rba	$abbrev
+rc	$only $abbrev
+rcc	$abbrev
+rcd	$abbrev
+rcs	$abbrev
+rd	$only $abbrev
+rds	$abbrev
+rgb	$abbrev
+rh	$only $abbrev
+rk	$only $abbrev
+rm	$only $abbrev
+rma $only $abbrev
+rme $only $abbrev
+rmf $only $abbrev
+rmi $only $abbrev
+rmt $only $abbrev
+rmx	$abbrev
+rn $only $abbrev
+rnk $only $abbrev
+rp	$only $abbrev
+rqy $only $abbrev
+rs	$abbrev
+rss	$abbrev
+rta	$abbrev
+rtf	$abbrev
+rtl	$abbrev
+rttl	$abbrev
+rtv	$abbrev
+rw	$abbrev
+rwo $only $abbrev
+rx	$abbrev
+sbr $only $abbrev
+scr	$only $abbrev
+sr	$only $abbrev
+srd	$only $abbrev
+ssa $only $abbrev
+sys $only $abbrev
+tpu $only $abbrev
+udf $only $abbrev
+udl $only $abbrev
+udt	$abbrev
+uhh $abbrev
+uif $only $abbrev
+uip $only $abbrev
+url	$only $abbrev
+usd	$only $abbrev
+usr	$only $abbrev
+uss	$only $abbrev
+uu $only $abbrev
+uue $only $abbrev
+uup $only $abbrev
+uxd $only $abbrev
+vba	$only $abbrev
+vbe $only $abbrev
+vbr	$abbrev
+vga	$abbrev
+wax $only $abbrev
+wbr	$abbrev
+wiz $only $abbrev
+wma $only $abbrev
+wri $only $abbrev
+wta	$abbrev
+xam $only $abbrev
+xba $only $abbrev
+xiw $only $abbrev
+xla $only $abbrev
+xpa $only $abbrev
+xpi $only $abbrev
+xsi $only $abbrev
+xxa $only $abbrev
+xxe $only $abbrev
+xy	$abbrev
+xyz	$abbrev
+yu $only $abbrev
+zaa $only $abbrev
+zoo $only $abbrev
+
+br	$dot $abbrev
+dr	$dot $abbrev
+mr	$dot $abbrev
+mrs	$dot $abbrev
+itd	$dot $abbrev
+npr	$dot $abbrev
+str	$only $dot $abbrev
+
+inc	$dot
+prof	$dot
+sl	$dot
+tj	$dot
+tzv	$dot
+
+// Demonstratives
+tko      $u+
+?4 šta      $u+
+što      $u+
+koga     $pause
+čega     $pause
+
+
+// Pronouns
+
+
+// Prepositions
+o        $u $brk
+u        $u $brk
+s        s_ $brk
+
+
+// Conjunctions
+a        $u $pause    // and
+i        $u $pause    // and
+ali      $pause       // but
+nego     $pause       // but
+ili      $u $pause    // or
+
+
+
+
+// Auxillary verbs
+li       $u      // question
+
+sam      $u      // to be, short form
+si       $u
+je       $u
+smo      $u
+ste      $u
+su       $u
+
+bio      $u       // to be, past tense
+bila     $u
+bilo     $u
+bili     $u
+
+nisam    $u+      // to be, negative
+nisi     $u+
+nije     $u+
+nismo    $u+
+niste    $u+
+nisu     $u+
+
+imam     $u+      // to have
+imaš     $u+
+ima      $u+
+imamo    $u+
+imate    $u+
+imaju    $u+
+
+nemam    $u+      // to have, negative
+nemaš    $u+
+nema     $u+
+nemamo   $u+
+nemate   $u+
+nemaju   $u+
+
+
+
+
+
+// treat as letters if at the end of the clause
+a        a    $atend    // letter-a
+i        i    $atend    // letter-i
+o        o    $atend    // letter-o
+s        s@   $atend
+u        u    $atend
+
+
+// Exceptions
+
+ctrl	tsontRo:l
+control	tsontRo:l
+docx	d'ots||'iks
+jpeg	j@2||p'eg
+macx	m'ats||'iks
+potx	p'ot||'iks
+sisx	s'is||'iks
+start	st'aRt
+utf	u||te||ef
+
+?2 ra	R&
+?2 re	Re
+?2 ar	&R
+?2 er	eR

dictsource/hr_rules → dictsource/hbs_rules

-// translation rules for Croatian
+
+// "hbs" is the ISO 639-3 code for Serbian/Croatian/Bosnian macrolanguage
 // This file is UTF-8 encoded
 
+//The next line gives a voice which is used to find a phoneme table to use to compile this dictionary
+//voice=hr
+
 // Conditional rules:
 // ?1  Croatian
-// ?2  Bosnian (Bosna and Hercegovina)
+// ?2  Serbian
+// ?3  Bosnian (Bosna and Hercegovina)
+// ?4  Serbian & Bosnian
 
 .group a
         a        a
         e        E
      j) e        e
         e (j     e
-    ._) exe      %ekse        // filetype
+     _) exe      %ekse        // filetype
+?2   _) exe      %eikse       // filetype
 
 .group f
         f        f
 .group q
         q        k
         qu       kv
+?2      qu       ku
 
 .group r
         r        R
      A) r (A     R
-     K) r (A     @-*
+     K) r (A     r*    // this phoneme expands to [@-*]
         rr       R
         r (_     R
 
         $        d'ol&R
 
         .        totSk&
-?2      .        tatSk&
+?4      .        tatSk&
         . (.     _:
    \.)  .
 
 	μ	m'i
 	ν	n'i
 	ξ	ks'i
-	ο	'omIk@-*on_
+	ο	'omIkr*on_
 
 .group 0xcf   // Greek letters
 
 	ψ	ps'i
 	ω	'omeg&
 	ώ	'omeg&||t'onos
-	ό	'omIk@-*on||t'onos
+	ό	'omIkr*on||t'onos
 
 

dictsource/hu_list

-// Translations for Hungarian words
-// This file is UTF-8 encoded.
-
-// Conditional rules:
-// ?1   speak "i" as Roman number
-
-
-// Numbers
-
-_0        nullA
-_1        EJ
-_2        kEttY:
-_3        ha:R2om
-_4        ne:J
-_5        Yt
-_6        hAt
-_7        he:t
-_8        n^olts
-_9        kilEnts
-_10       ti:z
-_1X       tizEn
-_20       hu:s
-_2X       huson
-_3X       hAR2mints
-_4X       nEJvEn
-_5X       YtvEn
-_6X       hAtvAn
-_7X       hEtvEn
-_8X       n^oltsvAn
-_9X       kilEntsvEn
-_0C       sa:z
-_1C       sa:z
-
-_0M1      EzER2
-_1M1      EzER2
-_2M1      ke:tEzER2
-_0M2      millio:
-_1M2      millio:
-_2M2      ke:tmillio:
-_0M3      millia:R2d
-_1M3      millia:R2d
-_1M3      ke:tmillia:R2d
-_0M4      billio:
-_1M4      billio:
-_2M4      ke:tbillio:
-_dpt      ||_vEss2Y:_
-
-
-// Abbreviations
-km	kilo:me:tER2
-mm      milime:tER2
-
-a.m	An^n^i||m'int  $dot
-dr	doktoR2  $dot
-gpu	$abbrev
-id	idY:SEb:  $dot
-ill	illEdvE  $dot
-ifj	ifjAb:  $dot
-mta	$abbrev
-pl	pe:lda:ul  $dot
-stb	SAtYb:i  $dot
-u.n	u:JnEvEzEt:   $dot
-ú.n	u:JnEvEzEt:   $dot
-vö	v'EZd||_'Yss2E
-
-?1 i	EJ	$abbrev     // option for Roman number
-
-
-// Articles
-a         $u     // the
-az        $u
-aki       $u+   // that
-
-ez        $u+   // this
-
-
-// Pronouns
-
-// If these are only used for emphasis then then don't need $u
-// én        $u
-// te        $u
-// ő         $u
-// mi        $u
-// ti        $u
-// ők        $u
-// maga      $u
-// maguk     $u
-
-
-// Conjunctions
-és        $u  $strend2 // and
-de        $u+ $pause  // but
-vagy      $u+ $pause  // or
-
-mert      $u $pause     // because
-miatt     $u           // because
-bár       $u $pause     // although
-habár     $u $pause     // although
-
-ha        $u $pause  // if
-noha      $u $pause  // even thought
-
-
-// Prepositions etc.
-
-
-// Negative
-nem       $alt2 $strend
-ne        $u+
-
-e         $u          // "-e" question
-e         $atend      // stressed at end of sentence
-
-// Unstressed adverbs prepositions etc.
-néhány    $u+
-is        $u         // also
-
-elöl	$u      // in front of
-elől	$u
-elé	$u
-előtt	$u
-
-mögül	$u      // behind
-mögött	$u
-mögé	$u
-
-kívül	$u
-után	$u     // after
-
-alól	$u
-alá	$u
-alatt	$u    // below
-
-felől	$u    // above
-felé	$u
-fölé	$u
-felett	$u    // above
-fölött	$u
-
-mellől	$u    // besides
-mellé	$u
-mellett	$u
-
-közül	$u
-közé	$u
-között	$u
-
-an	$u	// so many persons
-as	$u
-ba	$u	// into
-ban	$u	// in
-be	$u
-ben	$u
-ból	$u	// from inside
-ből	$u
-e       $u      // in query, his/her
-ei      $u      // his her multiole
-en	$u
-es	$u
-hez	$u
-hoz	$u      // to
-höz	$u
-ig	$u	// up to
-ik	$u      // -th
-nak     $u    // of, him/her
-nek     $u
-nál	$u	// by
-nél	$u
-on	$u	// on
-en	$u	// on
-os	$u
-ra	$u	// onto
-re	$u
-ról	$u      // from top of, about
-röl	$u
-ről	$u
-szer	$u
-szor	$u	// times
-szór	$u
-ször	$u
-tól	$u    // from
-töl	$u 
-től	$u
-val     $u    // with
-vel     $u
-án	$u    // on that day
-é	$u    // his, her
-éi	$u    // his, her multiple
-én	$u
-ón	$u
-ós	$u
-ön	$u
-ös	$u
-
-ellen	$u	// against
-elé	$u	// before
-óta	$u	// since
-
-sok	$u	// many
-kevés	$u	// few
-se	$u       // neither
-sem	$u       // nor
-
-
-be	$u	//into
-ki	$u	// out of
-le	$u	// down
-fel	$u	// up
-föl	$u	// up
-meg	$u	// finish an action
-el	$u	// away
-át	$u	//through
-rá	$u	//onto
-ide	$u	// here
-oda	$u	// there
-szét	$u	//apart
-össze	$u	//connect with
-vissza	$u	//back
-ellen	$u	//against
-
-volna	$u	// would
-részén	$u	//on ... part of
-
-nélkül	$u	// without
-folytán	$u	// because of
-úgy	$u	// like that
-így	$u	// like this
-egyaránt $u	// alike
-minden	$u	// all
-részben	$u	// partly
-
-egyik	$u	// one of
-másik	$u	// the other
-
-
-// word pairs
-
-(hogy a)   h,oJ,A
-(hogy az)  h,oJ,Az
-(hogy aki) h,oJ,Aki
-
-
-// Verbs
-
-vagyok    $u  // to be, present
-vagy      $u
-van       $u
-vagyunk   $u
-vagytok   $u
-vannak    $u
-
-voltam    $u  // to be, past
-voltál    $u
-volt      $u
-voltunk   $u
-voltatok  $u
-voltak    $u
-
-leszek    $u  // to be, future
-leszel    $u
-lesz      $u
-leszünk   $u
-lesztek   $u
-lesznek   $u
-
-
-kell      $u+  // should
-
-szabad    $u+  // can (allowed)
-
-fog       $u+  // going to
-fogok     $u+
-fogsz     $u+
-fogunk    $u+
-fogtok    $u+
-fognak    $u+
-
-tud       $u+  // can
-tudok     $u+
-tudsz     $u+
-tudunk    $u+
-tudtok    $u+
-tudnak    $u+
-
-// Letters
-_a	A:
-a	A: $atend     // letter "a"
-b	be:
-c	tse:
-d	de:
-_e	E
-f	Eff
-g	ge:
-h	ha:
-j	je:
-k	ka:
-l	Ell
-m	Emm
-n	Enn
-p	pe:
-q	ku
-r	ERR
-s	S_
-s	ES  $atend
-_s	ES
-v	ve:
-w	duplAve:
-x	iks
-y	ipsilon
-z	ze:
-
-_é	e:
-é	$atend
-
-cs	tSe:
-dz	dze:
-dzs	dZe:
-sz	Es
-zs	Ze:
-gy	Je:
-ty	ce:
-ly	Ellipsilon
-ny	En^
-
-_.	pont
-Ft	foR2int
-
-
-
-// Main Exceptions List
-//=====================
-
-nato	na:to:
-
-
-

espeak-data/voices/bs

 name bosnian
 language bs
 phonemes hr
-dictionary hr
+dictionary hbs
 gender male
 
 pitch 81 120

espeak-data/voices/grc

+name greek-ancient
+language grc
+gender male
+
+stressLength 170 170  190 190  0 0  230 240
+dictrules 1
+words 1
+

espeak-data/voices/hr

 name croatian
 language hr
+language hbs
 gender male
 
+dictionary hbs
+
 // attributes towards !variant3
 pitch 81 120
 formant 0 100 100 100

espeak-data/voices/sr

+name serbian
+language sr
+gender male
+dictionary hbs
+
+// attributes towards !variant3 pitch 80 120
+formant 0 100 100 100
+formant 1  97  97 100
+formant 2  97  97 100
+formant 3  97 102 100
+formant 4  97 102 100
+formant 5  97 102 100
+
+stressAdd 10 10 0 0 0 0 -30 -30

espeak-data/voices/zh

+name Mandarin test
+language zh
+gender male
+words 1
+
+//for some dialects
+
+//[en]: replace ng with n
+//[zh]: 无后鼻音，ng读成n
+//replace 0 N n
+
+//[en]: replace rfx consonants
+//[zh]: 无卷舌音，r读成l或z，er读成e
+//replace 0 ts.h tsh
+//replace 0 ts. ts
+//replace 0 s. s
+//replace 0 i. i[
+//replace 0 z. l
+//replace 0 z. z
+//replace 0 @r @
+
+//[en]: replace beginning n or l
+//[zh]: 不分nl，n读成l或l读成n
+//replace 2 n l
+//replace 2 l n
+
+//[en]: replace beginning w with v
+//[zh]: w读成v
+//replace 0 w  v

phsource/compile_report

       pl  15  107
       sk  29  125
       cs   6  125
-      hr  19  129
-      mk   3  130
-      sr   2  129
+      hr  20  130
+      mk   3  131
+      sr  18  130
       ru  38  126
       it  17  118
       la  21  114
                 [d;] ro
                 [d] el
                 [d] sw
+d/d_dnt         [d] sr
 d/dr            [*] base
                 [d] base
                 [d[] base
                 [dz;] base
                 [J] hi
                 [dZ;] mk
+                [dZ] sr
+                [dZ;] sr
                 [dZ;] ru
 dzh/dzh_        [dZ] base
                 [dZ;] base
 g/xg            [g] base
                 [g] cy
                 [g] es
+h/h_            [h] base
+                [h] fi
+                [H] hi
+                [h] hi
+                [<h>] la
+h/h@            [h] base
+                [h] fi
+                [H] hi
+                [h] hi
+                [th] hi
+                [h] sr
+                [<h>] la
+h/ha            [h] base
+                [h] fi
+                [H] hi
+                [h] hi
+                [<h>] la
+h/he            [h] base
+                [h] fi
+                [H] hi
+                [h] hi
+                [<h>] la
+h/hi            [h] base
+                [h] fi
+                [H] hi
+                [h] hi
+                [<h>] la
+h/ho            [h] base
+                [H] hi
+                [h] hi
+                [<h>] la
+h/hu            [h] base
+                [H] hi
+                [h] hi
+                [<h>] la
+h/hu_fi         [h] fi
 j2/_j2          [;] base
 j2/j2@          [(@)] base
 j2/j2a          [(a)] base
                 [(o)] base
                 [(u)] base
 m/_m            [m] base
+                [m] sr
                 [m;] ru
                 [m;] ro
                 [m] zh
 m/m_            [m] base
+                [m] sr
                 [m;] ru
                 [m;] ro
 m/m@            [(@)] base
 m/me            [(e)] base
 m/mi            [(i)] base
 m/mj            [m] base
+                [m] sr
                 [m;] ru
                 [m;] ro
 m/mo            [(o)] base
+m/m_sr.wav      [m] sr
 m/mu            [(u)] base
 n^/_n^          [n^] base
+                [n^] sr
 n^/n^_          [n^] base
+                [n^] sr
 n^/n^@          [(@)] base
 n^/n^#_         [n^#] is
 n/_n            [n] base
+                [n] sr
                 [n;] ro
                 [n] zh
 n/n_            [n] base
+                [n] sr
                 [n;] ro
                 [n] zh
 n/n@            [(@)] base
 n/nj            [n] base
                 [n.] base
                 [n^] base
+                [n] sr
+                [n^] sr
                 [n;] ro
+n^/nj_sr.wav    [N] sr
+                [n^] sr
 nn/_nn          [N] base
                 [N] en
+                [N] sr
                 [N] zh
 nn/nn_          [N] base
                 [N] en
+                [N] sr
                 [N] zh
 nn/nn@          [(@)] base
 nn/nn#_         [N#] is
 nn/nne          [(e)] base
 nn/nni          [(i)] base
 nn/nnj          [N] base
+                [N] sr
 nn/nno          [(o)] base
 nn/nnu          [(u)] base
 n^/n^o          [(o)] base
 n/no            [(o)] base
 n/_nr           [n.] base
 n/nr_           [n.] base
+n/n_sr.wav      [N] sr
+                [n] sr
 n^/n^u          [(u)] base
 n/nu            [(u)] base
 r2/_r2          [r-] base
 r3/r_n          [r/] base
                 [r/] ru
 r3/r_o          [(o)] base
+r3/r_sr.wav     [R] sr
+                [r*] sr
+                [r-] sr
 r3/r_trill      [R2] base
                 [R3] base
                 [r] af
                 [r] cy
                 [r] nl
                 [R] sk
+                [R] sr
+                [r*] sr
+                [r-] sr
                 [x] pt
 r3/r_trill2     [R] base
                 [R] hr
                 [(u)] base
 ufric/ch        [C] base
                 [C2] de
+ufric/ch_sr     [h] sr
+                [x] sr
 ufric/f         [f] base
                 [pF] de
                 [f;] ro
                 [f] zh
 ufric/f_        [f] base
                 [f;] ro
-ufric/h_        [h] base
-                [h] fi
-                [H] hi
-                [h] hi
-                [<h>] la
-ufric/h@        [h] base
-                [h] fi
-                [H] hi
-                [h] hi
-                [th] hi
-                [<h>] la
-ufric/ha        [h] base
-                [h] fi
-                [H] hi
-                [h] hi
-                [<h>] la
-ufric/he        [h] base
-                [h] fi
-                [H] hi
-                [h] hi
-                [<h>] la
-ufric/hi        [h] base
-                [h] fi
-                [H] hi
-                [h] hi
-                [<h>] la
-ufric/ho        [h] base
-                [H] hi
-                [h] hi
-                [<h>] la
-ufric/hu        [h] base
-                [H] hi
-                [h] hi
-                [<h>] la
-ufric/hu_fi     [h] fi
 ufric/ll        [L] base
                 [l#] is
 ufric/s         [s] base
 ufric/sh_pzd_   [S;] base
                 [Z;] base
 ufric/sh_rfx    [s.] base
+ufric/sh_sr.wav  [S] sr
+                [Z] sr
+                [dZ] sr
 ufric/s_pzd     [s;] base
                 [z;] base
                 [s;] ro
                 [x2] nl
                 [x2] pt
 ufric/x_hr      [x] hr
+ufric/x_sr      [h] sr
+                [x] sr
 ustop/c         [c] base
 ustop/k         [k] base
                 [k] en
                 [t] el
 ustop/tsh_pzd   [tS;] zh
 ustop/tsh_pzd_unasp  [tS;] zh
+ustop/tsh_sr.wav  [tS] sr
 ustop/ts_pzd    [tS;] base
                 [ts;] base
                 [ch] hi
                 [c] hu
 ustop/ts_pzd2   [c] hi
                 [c] hu
+ustop/t_sr      [d] sr
+                [t] sr
 ustop/ts_rfx    [ts.] zh
 ustop/ts_rfx_unasp  [ts.] zh
+ustop/ts_sr.wav  [tS;] sr
+                [dZ;] sr
 ustop/ts_unasp  [ts] zh
 ustop/t_unasp   [t] zh
 vdiph/0i        [OI] pt
 voc/zh          [Z] base
                 [z.] base
                 [R^] sk
+                [Z] sr
                 [Z;] ro
 voc/zh_         [Z] base
                 [z.] base
                 [@4] hi
                 [@] hr
 vowel/@-        [@-] base
+                [r*] hr
 vowel/&         [a] en_rp
                 [&] fi
                 [&:] hi
                 [a:] sk
                 [A] hr
                 [a] hr
-                [a] sr
                 [a] it
                 [a:] la
                 [a] zh

phsource/ph_croatian

   lengthmod 6
 endphoneme
 
+phoneme r*
+  vowel nonsyllabic
+  starttype (@)  endtype (@)
+  formants vowel/@-
+  unstressed
+  appendph *
+  length 50
+endphoneme
 

phsource/ph_finnish

   vls glt apr
   length 100
   lengthmod 3
-  wave ufric/h_%70         // no vowel following
-  before (@) ufric/h@
-  before (a) ufric/ha
-  before (e) ufric/he
-  before (i) ufric/hi
-  before (o) ufric/hu_fi%70
-  before (u) ufric/hu_fi%80
+  wave h/h_%70         // no vowel following
+  before (@) h/h@
+  before (a) h/ha
+  before (e) h/he
+  before (i) h/hi
+  before (o) h/hu_fi%70
+  before (u) h/hu_fi%80
 endphoneme
 

phsource/ph_hindi

   vls glt apr
   length 100
   lengthmod 3
-  wave ufric/h_%70         // no vowel following
-  before (@) ufric/h@+35
-  before (a) ufric/ha+35
-  before (e) ufric/he+35
-  before (i) ufric/hi+35
-  before (o) ufric/ho+35
-  before (u) ufric/hu+35
+  wave h/h_%70         // no vowel following
+  before (@) h/h@+35
+  before (a) h/ha+35
+  before (e) h/he+35
+  before (i) h/hi+35
+  before (o) h/ho+35
+  before (u) h/hu+35
 endphoneme
 
 phoneme H       // use for full "h"
   vls glt apr
   length 100
   lengthmod 3
-  wave ufric/h_%70         // no vowel following
-  before (@) ufric/h@
-  before (a) ufric/ha
-  before (e) ufric/he
-  before (i) ufric/hi
-  before (o) ufric/ho
-  before (u) ufric/hu
+  wave h/h_%70         // no vowel following
+  before (@) h/h@
+  before (a) h/ha
+  before (e) h/he
+  before (i) h/hi
+  before (o) h/ho
+  before (u) h/hu
 endphoneme
 
 
   vowelin f1=0  f2=1500 -300 300  f3=-100 80
   vowelout f1=0 f2=1500 -300 250  f3=-100 80  rms=20
   lengthmod 2
-  wave  ustop/t_hi%50+ufric/h@
+  wave  ustop/t_hi%50+h/h@
   switchvoicing d
 endphoneme
 

phsource/ph_latin

   vls glt apr
   length 50
   lengthmod 3
-  wave ufric/h_%70         // no vowel following
-  before (@) ufric/h@
-  before (a) ufric/ha
-  before (e) ufric/he
-  before (i) ufric/hi
-  before (o) ufric/ho
-  before (u) ufric/hu
+  wave h/h_%70         // no vowel following
+  before (@) h/h@
+  before (a) h/ha
+  before (e) h/he
+  before (i) h/hi
+  before (o) h/ho
+  before (u) h/hu
 endphoneme
 
 

phsource/ph_serbian

-// include any phonemes which are different from those
-// which are inherited from the base language(s)
-
-phoneme a
-  vowel starttype (a) endtype (a)
-  length 160
-  formants vowel/a
-  reduceto &  4
+// ufric/x_sr
+// ufric/ch_sr
+// ufric/sh_sr
+// ustop/t_sr
+// ustop/ts_sr
+// ustop/tsh_sr
+// r3/r_sr
+// m/m_sr
+// n/n_sr
+// n^/nj_sr
+
+// This file inherits phonemes from ph_croatian
+// Only phonemes which differ from the inherited phonemes need
+// to be included here.
+
+
+phoneme t        // dental variant of /t/
+  vls dnt stop
+  vowelin f1=0  f2=1600 -300 300  f3=-100 80
+  vowelout f1=0 f2=1600 -300 250  f3=-100 80  rms=20
+  lengthmod 2
+  wave  ustop/t_sr%50
+  before _ ustop/t_sr%35
+  switchvoicing d
+endphoneme
+
+
+phoneme d
+  vcd alv stop
+  vowelin f1=2  f2=1700 -300 300  f3=-100 80 amp=16
+  vowelout f1=2 f2=1700 -300 300  f3=-100 80
+  formants d/d_dnt+ustop/t_sr%50
+  before _ d/d_dnt+ustop/t_sr
+  before r d/d_dnt+ustop/t_sr%50
+  after _ d/d_dnt+ustop/t_sr
+  after @ d/d_dnt+ustop/t_sr
+  lengthmod 5
+  switchvoicing t
+endphoneme
+
+
+// This is for letter "h" at the beginning of a word
+phoneme x
+  vls vel frc
+  lengthmod 3
+  wave ufric/x_sr%70         // no vowel following
+  before (a) ufric/x_sr
+  before (e) ufric/ch_sr
+  before (i) ufric/ch_sr
+  before (o) ufric/x_sr
+  before (u) ufric/x_sr
+endphoneme
+
+
+// This is for letter "h" not at the beginning of  word
+phoneme h
+  vls vel frc
+  lengthmod 3
+  wave ufric/x_sr%70         // no vowel following
+  before (@) h/h@
+  before (a) ufric/x_sr
+  before (e) ufric/ch_sr
+  before (i) ufric/ch_sr
+  before (o) ufric/x_sr
+  before (u) ufric/x_sr
+endphoneme
+
+
+
+phoneme R
+  liquid
+  vowelin  f1=0  f2=1600 -300 300  f3=-200 80
+  vowelout f1=2  f2=1600 -300 300  f3=-200 80 brk
+  formants r3/r_trill+r3/r_sr.wav%65
+  after & r3/r_trill+r3/r_sr.wav
+  after e r3/r_trill+r3/r_sr.wav
+  after i r3/r_trill+r3/r_sr.wav
+  after o r3/r_trill+r3/r_sr.wav
+  after u r3/r_trill+r3/r_sr.wav
+  after y r3/r_trill+r3/r_sr.wav
+  trill
+  lengthmod 6
+endphoneme
+
+// This is "r" which doesn't follow a vowel
+phoneme r*
+  liquid
+  vowelin  f1=0  f2=1600 -300 300  f3=-200 80
+  vowelout f1=2  f2=1600 -300 300  f3=-200 80 brk
+  formants r3/r_trill+r3/r_sr.wav%65
+  trill
+  lengthmod 6
+endphoneme
+
+// This is "r" which doesn't have a vowel before or after it
+phoneme r-
+  liquid
+  vowelin  f1=0  f2=1600 -300 300  f3=-200 80
+  vowelout f1=2  f2=1600 -300 300  f3=-200 80 brk
+  formants r3/r_trill+r3/r_sr.wav%65
+  trill
+  lengthmod 6
+endphoneme
+
+
+
+phoneme dZ
+  vcd pla afr sibilant
+  vowelin f1=2  f2=2300 200 400  f3=100 80
+  vowelout f1=2  f2=2300 250 300  f3=100 80 brk
+  formants dzh/dzh+ufric/sh_sr.wav
+  before _ dzh/dzh+ufric/sh_sr.wav
+  after _ NULL
+  after @ dzh/dzh+ufric/sh_sr.wav
+  lengthmod 5
+  switchvoicing tS
+endphoneme
+
+
+phoneme dZ;
+  vcd pla afr sibilant palatal
+  vowelin f1=2  f2=2700 400 600  f3=300 80 rate len=70
+  formants dzh/dzh+ustop/ts_sr.wav
+  before _ dzh/dzh+ustop/ts_sr.wav
+  after _ NULL
+  after @ dzh/dzh+ustop/ts_sr.wav
+  lengthmod 5
+  switchvoicing tS;
+endphoneme
+
+
+phoneme tS
+  vls pla afr sibilant
+  vowelin f1=0  f2=2300 200 400  f3=-100 80
+  lengthmod 2
+  wave ustop/tsh_sr.wav
+  before _ ustop/tsh_sr.wav
+  switchvoicing dZ
+endphoneme
+
+
+phoneme tS;
+  vls pal afr sibilant palatal
+  vowelin f1=0  f2=2700 400 600  f3=300 80 rate len=70
+  lengthmod 2
+  wave ustop/ts_sr.wav
+  before _ ustop/ts_sr.wav
+  switchvoicing dZ;
+endphoneme
+
+
+phoneme Z
+  vcd pla frc sibilant
+  vowelin  f1=0  f2=2000 0 300  f3=-200 80
+  vowelout f1=2  f2=2000 0 300  f3=-200 80 brk
+  formants voc/zh+ufric/sh_sr.wav
+  before _ voc/zh+ufric/sh_sr.wav
+  lengthmod 6
+  switchvoicing S
+endphoneme
+
+
+phoneme S
+  vls pla frc sibilant
+  vowelin  f1=0  f2=2100 100 300  f3=-100 80
+  vowelout f1=0  f2=2100 100 300  f3=-100 80
+  length 100
+  lengthmod 3
+  wave ufric/sh_sr.wav
+  before _ ufric/sh_sr.wav
+  switchvoicing Z
+endphoneme
+
+
+phoneme m
+  vcd blb nasal
+  vowelout f1=2  f2=1000 -500 -350  f3=-200 80 brk
+  length 100
+  lengthmod 4
+  formants NULL
+  after _ m/_m+m/m_sr.wav
+  before _ m/m_+m/m_sr.wav
+  before (r) m/mj+m/m_sr.wav
+endphoneme
+
+
+phoneme n
+  vcd alv nasal
+  vowelout f1=2 f2=1700 -300 250  f3=-100 80  rms=20 brk
+  length 100
+  lengthmod 4
+  formants NULL
+  after _ n/_n+n/n_sr.wav
+  after n n/_n+n/n_sr.wav      // lengthen [nn]
+  before _ n/n_+n/n_sr.wav
+  before (r) n/nj+n/n_sr.wav
+endphoneme
+
+
+phoneme n^
+  vcd pal nasal palatal
+  vowelout f1=3 f2=2500 300 500  f3=100 80  len=45 brk rate
+  length 100
+  lengthmod 4
+  formants NULL
+  after _ n^/_n^+n^/nj_sr.wav
+  before _ n^/n^_+n^/nj_sr.wav
+  before (r) n/nj+n^/nj_sr.wav
+endphoneme
+
+
+phoneme N
+  vcd vel nasal
+//  vowelout  brk
+  vowelout f1=3 f2=2300 300 400  f3=-200 80  len=40 rms=20 brk
+  length 100
+  formants NULL
+  after _ nn/_nn+n/n_sr.wav
+  before _ nn/nn_+n/n_sr.wav
+  before (r) nn/nnj+n^/nj_sr.wav
+  lengthmod 8            // longer than 'n'
 endphoneme

phsource/phonemes

 phonemenumber 21 _^_    //  Change language
 phonemenumber 22 _X1    // a language specific action
 phonemenumber 23 _|     // very short pause
-phonemenumber 24 t#     // reduced version of [t]
+phonemenumber 24 _::    // long pause
+phonemenumber 25 t#     // reduced version of [t]
 
 // The base phoneme table contains the utility phonemes, the consonants,
 // and a schwa.  The vowels are defined in the language specific phoneme
   starttype _ endtype _
   length 45
   lengthmod 5   // ??
+  beforenotvowel _  // change to pause_short
   nolink
 endphoneme
 
   vls glt apr
   length 100
   lengthmod 3
-  wave ufric/h_%70         // no vowel following
-  before (@) ufric/h@
-  before (a) ufric/ha
-  before (e) ufric/he
-  before (i) ufric/hi
-  before (o) ufric/ho
-  before (u) ufric/hu
+  wave h/h_%70         // no vowel following
+  before (@) h/h@
+  before (a) h/ha
+  before (e) h/he
+  before (i) h/hi
+  before (o) h/ho
+  before (u) h/hu
 endphoneme
 
 

src/compiledata.cpp

 	int err;
 	int errors = 0;
 	int dict_count = 0;
+	FILE *f_in;
 	FILE *log;
 	FILE *f_phused;
 	char dictname[80];
 	char fname_log[sizeof(path_dsource)+20];
 	char save_voice_name[80];
+	char path[sizeof(path_home)+40];       // path_dsource+20
+	char buf[200];
+	char voicename[80];
 
 	if(!wxDirExists(path_dictsource))
 	{
 		dictstr = fname.GetName().BeforeLast('_');
 		strcpy(dictname,dictstr.mb_str(wxConvLocal));
 		dict_count++;
+		strcpy(voicename,dictname);
+
+		// read the *_rules file to see if a phoneme table is specified though a voice name
+		sprintf(path,"%s%s_rules",path_dsource,dictname);
+		if((f_in = fopen(path,"r")) != NULL)
+		{
+			unsigned int ix;
+			unsigned int c;
+
+			for(ix=0; ix<20; ix++)
+			{
+				if(fgets(buf,sizeof(buf),f_in) == NULL)
+					break;
+
+				if(memcmp(buf,"//voice=",8)==0)
+				{
+					for(ix=0; ix<sizeof(voicename); ix++)
+					{
+						if(isspace(c = buf[ix+8]))
+						{
+							break;
+						}
+						voicename[ix] = c;
+					}
+					voicename[ix] = 0;
+					break;
+				}
+			}
+			fclose(f_in);
+		}
 
-		LoadVoice(dictname,0);
+		LoadVoice(voicename,0);
 
 		if((err = CompileDictionary(path_dsource, dictname,log,NULL)) > 0)
 		{

src/compiledict.cpp

 //#define OPT_FORMAT         // format the text and write formatted copy to Log file 
 //#define OUTPUT_FORMAT
 
+extern void Write4Bytes(FILE *f, int value);
 int HashDictionary(const char *string);
 
 static FILE *f_log = NULL;
 				}
 				if(replace1 != 0)
 				{
-					fwrite(&replace1,1,4,f_out);
-					fwrite(&replace2,1,4,f_out);
+					Write4Bytes(f_out,replace1);   // write as little-endian
+					Write4Bytes(f_out,replace2);   // if big-endian, reverse the bytes in LoadDictionary()
 				}
 			}
 			break;

src/dictionary.cpp

 	to[size-1] = 0;
 }
 
+#ifdef ARCH_BIG
+int reverse_word_bytes(int word)
+{//=============================
+	// reverse the order of bytes from little-endian to big-endian
+	int ix;
+	int word2 = 0;
+
+	for(ix=24; ix>=0; ix -= 8)
+	{
+		word2 |= (word >> ix) & 0xff;
+		word2 = word2 << 8;
+	}
+	return(word2);
+}
+#endif
 
 
 int LookupMnem(MNEM_TAB *table, char *string)
 				pw += 2;   // find the end of the replacement list, each entry is 2 words.
 			}
 			p = (char *)(pw+1);
+
+#ifdef ARCH_BIG
+			pw = (unsigned int *)langopts.replace_chars;
+			while(*pw != 0)
+			{
+				*pw = reverse_word_bytes(*pw);
+				pw++;
+			}
+#endif
 			continue;
 		}
 
 
 	if((ph = phoneme_tab[*p]) != NULL)
 	{
-		int gap = langopts.word_gap & 0xf00;
 
 		if(ph->type == phSTRESS)
 			ph = phoneme_tab[p[1]];
 
-		if(((gap) && (vowel_stress[1] >= 4) && (prev_stress >= 4)) || (langopts.word_gap & 0x8))
+#ifdef deleted
+		int gap = langopts.word_gap & 0x700;
+		if((gap) && (vowel_stress[1] >= 4) && (prev_stress >= 4))
 		{
 			/* two primary stresses together, insert a short pause */
-			if(gap == 0x0300)
-				*output++ = phonPAUSE;
-			else
-			if((gap == 0x0200) && (ph->type == phVOWEL))
-				*output++ = phonGLOTTALSTOP;
-			else
-				*output++ = phonPAUSE_SHORT;
+			*output++ = pause_phonemes[gap >> 8];
 		}
 		else
+#endif
 		if((langopts.vowel_pause & 0x30) && (ph->type == phVOWEL))
 		{
 			// word starts with a vowel

src/phoneme.h

 #define phonSWITCH      21
 #define phonX1          22      // a language specific action
 #define phonPAUSE_VSHORT 23
-#define phonT_REDUCED   24
-
+#define phonPAUSE_LONG  24
+#define phonT_REDUCED   25
+extern const unsigned char pause_phonemes[8];  // 0, vshort, short, pause, long, glottalstop
 
 // place of articulation
 #define phPLACE        0xf0000

src/phonemelist.cpp

 #include "translate.h"
 
 
+const unsigned char pause_phonemes[8] = {0, phonPAUSE_VSHORT, phonPAUSE_SHORT, phonPAUSE, phonPAUSE_LONG, phonGLOTTALSTOP, phonPAUSE_LONG, phonPAUSE_LONG};
 
 
 int Translator::ChangePhonemes(PHONEME_LIST2 *phlist, int n_ph, int index, PHONEME_TAB *ph, CHANGEPH *ch)
 		{
 			int x;
 
-			if(langopts.word_gap & 1)
-			{
-				insert_ph = phonPAUSE_VSHORT;
-			}
 			if(langopts.vowel_pause && (ph->type != phPAUSE) && (next->type == phVOWEL))
 			{
 				if(langopts.vowel_pause & 0x04)
 					insert_ph = phonPAUSE_SHORT;
 				}
 			}
+
+			if((x = (langopts.word_gap & 0x7)) != 0)
+			{
+				insert_ph = pause_phonemes[x];
+			}
 		}
 
 		next2 = phoneme_tab[(plist2+2)->phcode];
 				}
 			}
 			else
-			if(((langopts.word_gap & 2)==0) || ((plist2+1)->sourceix == 0))
+			if(((langopts.word_gap & 8)==0) || ((plist2+1)->sourceix == 0))
 			{
 				// This phoneme can be linked to a following vowel by inserting a linking phoneme
 				if(next->type == phVOWEL)

src/setlengths.cpp

 			{
 				len = langopts.length_mods0[next2->ph->length_mod *10+ next->ph->length_mod];
 
-				if((next->newword) && (langopts.word_gap & 0x4))
+				if((next->newword) && (langopts.word_gap & 0x20))
 				{
 					// consider as a pause + first phoneme of the next word
 					length_mod = (len + langopts.length_mods0[next->ph->length_mod *10+ 1])/2;

src/speak.cpp

 
 
 
-static void init_path(void)
-{//========================
+static void init_path(char *argv0)
+{//===============================
 #ifdef PLATFORM_WINDOWS
 	HKEY RegKey;
 	unsigned long size;
 	unsigned long var_type;
+	char *p;
 	char *env;
 	unsigned char buf[sizeof(path_home)-12];
 
 	{
 		sprintf(path_home,"%s\\espeak-data",env);
 		if(GetFileLength(path_home) == -2)
-			return;   // an espeak-data directory exists 
+			return;   // an espeak-data directory exists in the directory specified by environment variable
+	}
+
+	strcpy(path_home,argv0);
+	if((p = strrchr(path_home,'\\')) != NULL)
+	{
+		strcpy(&p[1],"espeak-data");
+		if(GetFileLength(path_home) == -2)
+			return;   // an espeak-data directory exists in the same directory as the espeak program
 	}
 
+	// otherwise, look in the Windows Registry
 	buf[0] = 0;
 	RegOpenKeyEx(HKEY_LOCAL_MACHINE, "Software\\Microsoft\\Speech\\Voices\\Tokens\\eSpeak", 0, KEY_READ, &RegKey);
 	size = sizeof(buf);
 	option_multibyte = espeakCHARS_AUTO;  // auto
 	f_trans = stdout;
 
-	init_path();
+	init_path(argv[0]);
 
 #ifdef NEED_GETOPT
 	optind = 1;

src/speech.h

 int LookupMnem(MNEM_TAB *table, char *string);
 
 
+#ifdef PLATFORM_WINDOWS
+#define N_PATH_HOME  220
+#else
+#define N_PATH_HOME  150
+#endif
 
-#define N_PATH_HOME  120
 extern char path_home[N_PATH_HOME];    // this is the espeak-data directory
 
 extern void strncpy0(char *to,const char *from, int size);

src/synthdata.cpp

 #include "translate.h"
 #include "wave.h"
 
-const char *version_string = "1.29.17  08.Nov.07";
+const char *version_string = "1.29.18  10.Nov.07";
 const int version_phdata  = 0x012901;
 
 int option_device_number = -1;

src/synthesize.cpp

 				DoMarker(espeakEVENT_WORD, (p->sourceix & 0x7ff) + clause_start_char, p->sourceix >> 11, clause_start_word + word_count++);
 		}
 
-		if((translator->langopts.word_gap & 1) || (translator->langopts.vowel_pause && (next->type == phVOWEL)))
-		{
-			// prevent word merging into next, make it look as though next is a pause
-			if((next->newword) && (next->type != phPAUSE))
-			{
-//				next_pause.ph = phoneme_tab[phonPAUSE];
-//				next = &next_pause;
-			}
-		}
-
 		EndAmplitude();
 
 		if(p->prepause > 0)

src/tr_english.cpp

 	static const short stress_lengths2[8] = {182,140, 220,220, 0,0, 248,275};
 
 	memcpy(stress_lengths,stress_lengths2,sizeof(stress_lengths));
-	langopts.vowel_pause = 0;
 	langopts.stress_rule = 0;
-	langopts.word_gap = 0;
 
 	langopts.numbers = 0x41 + NUM_ROMAN;
 }

src/tr_languages.cpp

 
 			// 'diminished' is an unstressed final syllable
 			tr->langopts.stress_flags =  0x6 | 0x10; 
-			tr->langopts.word_gap = 0;
-			tr->langopts.vowel_pause = 0;
 			tr->langopts.unstressed_wd1 = 0;
 			tr->langopts.unstressed_wd2 = 2;
 			tr->langopts.param[LOPT_SONORANT_MIN] = 120;  // limit the shortening of sonorants before short vowels
 			static const short stress_lengths_de[8] = {150,130, 190,190,  0, 0, 260,275};
 			tr = new Translator();
 			tr->langopts.stress_rule = 0;
-			tr->langopts.word_gap = 0x2;   // don't use linking phonemes
+			tr->langopts.word_gap = 0x8;   // don't use linking phonemes
 			tr->langopts.vowel_pause = 0x30;
 			tr->langopts.param[LOPT_PREFIXES] = 1;
 			memcpy(tr->stress_lengths,stress_lengths_de,sizeof(tr->stress_lengths));
 			tr->langopts.length_mods0 = tr->langopts.length_mods;  // don't lengthen vowels in the last syllable
 			tr->langopts.stress_rule = 2;
 			tr->langopts.stress_flags = 0x6;  // mark unstressed final syllables as diminished
-			tr->langopts.word_gap = 0;
-			tr->langopts.vowel_pause = 0;
 			tr->langopts.unstressed_wd1 = 0;
 			tr->langopts.unstressed_wd2 = 2;
 			tr->langopts.param[LOPT_SONORANT_MIN] = 130;  // limit the shortening of sonorants before short vowels
 			// stress last syllable if it doesn't end in vowel or "s" or "n"
 			// 'diminished' is an unstressed final syllable
 			tr->langopts.stress_flags = 0x200 | 0x6 | 0x10; 
-			tr->langopts.word_gap = 0;
-			tr->langopts.vowel_pause = 0;
 			tr->langopts.unstressed_wd1 = 0;
 			tr->langopts.unstressed_wd2 = 2;
 			tr->langopts.param[LOPT_SONORANT_MIN] = 120;  // limit the shortening of sonorants before short vowels
 
 			tr->langopts.length_mods0 = tr->langopts.length_mods;  // don't lengthen vowels in the last syllable
 			tr->langopts.stress_rule = 2;
-			tr->langopts.word_gap = 0;
 			tr->langopts.vowel_pause = 1;
 			tr->langopts.unstressed_wd1 = 2;
 			tr->langopts.unstressed_wd2 = 2;
 			tr = new Translator();
 
 			tr->langopts.stress_rule = 0;
-			tr->langopts.word_gap = 0;
 			tr->langopts.vowel_pause = 1;
 			tr->langopts.param[LOPT_DIERESES] = 1;
 			tr->langopts.param[LOPT_PREFIXES] = 1;
 			tr->langopts.length_mods0 = tr->langopts.length_mods;  // don't lengthen vowels in the last syllable
 
 			tr->langopts.stress_rule = 0;
-			tr->langopts.word_gap = 0x4;   // length of a final vowel is less dependent on the next consonant
+			tr->langopts.word_gap = 0x20;   // length of a final vowel is less dependent on the next consonant
 			tr->langopts.vowel_pause = 4;
 			tr->letter_groups[0] = vowels_vi;
 //			tr->langopts.param[LOPT_UNPRONOUNCABLE] = 1;   // disable check for unpronouncable words
 	case L('z','h'):
 	case L_zhy:
 		{
-			static const short stress_lengths_zh[8] = {220,150, 230,230, 230,0, 230,250};
+			static const short stress_lengths_zh[8] = {230,150, 230,230, 230,0, 230,250};
 			static const unsigned char stress_amps_zh[] = {22,16, 22,22, 22,22, 22,22 };
 
 			tr = new Translator;
 			tr->langopts.length_mods0 = tr->langopts.length_mods;  // don't lengthen vowels in the last syllable
 			tr->langopts.tone_numbers = 1;   // a number after letters indicates a tone number (eg. pinyin or jyutping)
 			tr->langopts.ideographs = 1;
-			tr->langopts.word_gap = 0x5;   // length of a final vowel is less dependent on the next consonant, don't merge consonant with next word
+			tr->langopts.word_gap = 0x21;   // length of a final vowel is less dependent on the next consonant, don't merge consonant with next word
 			if(name2 == L('z','h'))
 			{
 				tr->langopts.textmode = 1;
 
 	static const short stress_lengths2[8] = {170,140, 220,220,  0, 0, 250,270};
 	langopts.stress_rule = 0;
-	langopts.word_gap = 0;
 	langopts.vowel_pause = 0x30;
 	langopts.param[LOPT_DIERESES] = 1;
 	langopts.param[LOPT_PREFIXES] = 1;

src/translate.h

 
 
 typedef struct {
-// bit0=don't link consonants with the next word
-// bit1=don't add linking phonemes
-// bit2=length of a final vowel doesn't depend on the next phoneme
+// bits0-2  separate words with (1=pause_vshort, 2=pause_short, 3=pause, 4=pause_long 5=[?] phonemme)
+// bit 3=don't use linking phoneme
 // bit4=longer pause before STOP, VSTOP,FRIC
-// bit8-11=separate words with adjacent stressed syllables with (1=pause, 2=long pausem 3=[?] phoneme)
+// bit5=length of a final vowel doesn't depend on the next phoneme
 	int word_gap;
 	int vowel_pause;
 	int stress_rule; // 1=first syllable, 2=penultimate,  3=last