[1.27.06] Roman number translation added (for some languages).

espeak_Char(32) can speak "space" by adding an entry for _#32 in *_list file. Similarly for Tab.


git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@61 d46cf337-b52f-0410-862d-fd96e6ae7743

dictsource/af_list

-//maintainer: Willem van der Walt
+//maintainer: Willem van der Walt
 // email: [email protected]
 // tel: +27 12 8413996
 // Meraka Institute, CSIR
 samelewing    sA:m@l,e@v@N
 sedert        se@d@rt
 sektore       $2
+sinteties $2
 skepsele      skEps@l@
 s'n           s'@n
 sobere s'o@b@r@

dictsource/af_rules

              a (C         a
           C) a (Csier     %a // (ge)finansier/kassier
           j) a (pan       %a //japan first a sound and stress
+
           C) abie        'A:bi
              abo (C       %abO // move default stress and fix a and o sounds: abominasie/abonnement/aborsie
           _) a (bC        %a // abdy/ablatief/ablusie/abstrak, etdc.
           _) aka          a%k@ //akademie etc.
              akkedis      a%k@d'Is //akkedis and compounds
              ae           A:       //should be A:@ but sounds wrong
+          konst) abel A:b@l //exception for konstabel in compounds
+          t) abel %ab'&l //tabelopskrif temperatuurtabel etc.
+          t) abell %ab'&l //tabelle fix double l 
+          w) a (tering A: //watering exception to the rest
+          C) a (tering %a //short unstressed a before tering formatering etc. except watering
              ai           aI
              adres        %adrEs // fix last e sound of adres as last part of compounds
           _) alle (C      al@ // alledaags/e/allerhande
        CoeC) e (C         @ //fix e in boetedoening, boedelbelasting and many others
         orr) e (l         @ //fix e in borrel, orrel, korrel  in compounds
      intell) e (k        %E //intellek, and friends
-
+          p) e (ring 'e@ //imansipering and others e sound and stress
+          roep) e (ring 'e@ //groepering e sound and stress
              eval (u      %iv%al // evaluasie/evalueer/evaluering
         CAt) e (gn        & //fix e in mikrotegnologie and others
          ei) e            @ //eiewys and others
              iber (i@     %ibe@r // Liberië/Siberië/Liberiese/Siberiese/Iberiese
              ie           i
           h) i (dro       i //hidro in compounds
+          _) in (a@ In_ //inasem inagneming etc.
           _) iese (C      ,is@ // iesegrimmig and derivitives
              ieus         i'Y@s // this ending always takes stress
           C) i (g         @// figuur/figure/stoksielsaligalleenA
              minder (jar  m@n@r // move default stress: minderjarig and derivitives
              miner (A     m@n@r // mineraal and many compounds
              mi (neur     m%i // move default stress: mineur and compounds
-             minnares (_  m@n%arEs // fix stress and e sound: minnares/medeminnares and similar
              mikr (o      mikr
              ministers    m@n'Ist@rs
              mise (ra     m@s // fix stress and drop e: miserabel/-e/-s/-ste
           _) na (skeer    n''A: // restore default stress: naskeermiddel/-room
           _) na (tC       na // many compounds beginning with nat
              na (tuur     n%a // natuur and compounds and derivitives
+             navig n%av%ix2 //navigasie, navigeer etc.
           @) na (_        na // shorten final na, except for:
           r) na (_        nA: // daarna/hierna/waarna
              nadruklik    nA:dr'Wkl@k // nadruklik/e
              te (oo       t@ // geboorteoorskot/kosteoorweging/siekteoorsaak/terroristeoorlog
           _) teraarde     t%&r?A:rd@ // move default stress and fix e sounds: teraardebestelling/-s
           n) terie        t@r'i // galanterie/infanterie and compounds
+          wa) te (ring t@ //water exception to rest
+          Ca) te (ring t'e@ //opdatering formatering etc.
           _) tering       te@r@N // tering and compounds
              tesou (rie   t%Es%u //tesourie and compounds
       moles) tering       te@r@N // molestering/kindermolestering
           _) ter (l       t@r // move default stress: terloops/-e
           @) term (_      t&r@ // correct pronunciation of e
           @) terme (_     t&rm@ // correct pronunciation of 1st e
+          sin) te (ti te@ //sintetiseer, compounds and derivitives
              tern (K      t'&:r@-n //lantern ekstern etc.
              tern (e_     t'&rn // interne/eksterne
           _) ter (nA      t@r // terneergedruk/ternouernood
              ven (ster    fe~n // nasalise
              ven (yn      f@n // venyn/venynig
              ver          f@r
+             virtu v@rtS%y //virtueel virtuose etc. fix v and t sounds and assure lack of stress
           _) ver (@P3     f@r
              verant (w    f@r%ant // (on)verantwoordelik/-heid/verantwoording
              verende      f'e@r@nd@
              woe (styn    v%u // move default stress: woestyn and compounds
              wolke        vOlk@ // nimbuswolke/wolkekrabber
              woorde       vo@rd@
-
+             w (yser v //padwyser and many others with wyser in compounds
           nt) w (A v
 
 .group x
              ** 
         **)  * (_         _::
 
-             +            plus
+             +            pl'Ws
          _)  ++ (_        pluspl'us
     \+\+\+)  +          // ignore + after the first 3
 

dictsource/dict_phonemes

 *    *;   b    b;   c    C    d    d;   
 dZ   f    f;   g    h    j    k    l    
 l;   m    m;   n    N    n;   p    p;   
-r    s    S    s;   S;   t    t;   tS   
-ts   ts;  v    v;   w    z    Z    z;   
-Z;   
+r    s    S    S;   t    T    t;   tS   
+ts   ts;  v    v;   w    w2   x    z    
+Z    z;   Z;   
 
 
 Dictionary is_dict

dictsource/en_list

 _cap k,ap@-t@L
 _?A  lEt3
 _??  sImb@L
+_#9  tab
+_#32 speIs
 
 _!   Ekskl@m'eIS@n
 _"   kwoUts
 _0and	@n
 ?3 _0and
 _dpt	pOInt
+_roman	roUm@n_
 
 
 // ABBREVIATIONS
 wwii	dVb@Lju:dVb@Lju:t'u:
 xy	$abbrev
 
-ii	tu:       $abbrev    // roman numerals
-iii	Tri:      $abbrev
-iv	fo@       $abbrev
-
 
 nd                $only
 rd                $only

dictsource/es_list

 
 
 // numbers
-_0       TE*o
-_1       'uno
-_2       d'os
-_3       t@-*'es
-_4       kw'at@-*o
-_5       T'inko
-_6       s'eIs
-_7       sj'Ete
-_8       'otSo
-_9       nw'Eve
-_1X      dj'ET
-_11      'onTe
-_12      d'oTe
-_13      t@-*'ETe
-_14      kat'oRTe
-_15      k'inTe
-_20      v'eInte
-_2X      v'eInt
-_3X      t@-*'eInta
-_4X      kwa*'Enta
-_5X      Tinkw'Enta
-_6X      sEs'Enta
-_7X      sEt'Enta
-_8X      otS'Enta
-_9X      nov'Enta
-_0C      T'ientos
-_1C0     T'ien         // exactly one hundred
-_1C      T'iento
-_5C      kinj'Entos
-_0M1     m'il
-_1M1     m'il          // no '1' before thousand
-_0M2     mil^'ones
-_1M2     'unmil^'on
-_0M4     _bil^'ones
-_1M4     'unbil^'on
-_0and    i
-_dpt     koma
+_0	TE*o
+_1	'uno
+_2	d'os
+_3	t@-*'es
+_4	kw'at@-*o
+_5	T'inko
+_6	s'eIs
+_7	sj'Ete
+_8	'otSo
+_9	nw'Eve
+_1X	dj'ET
+_11	'onTe
+_12	d'oTe
+_13	t@-*'ETe
+_14	kat'oRTe
+_15	k'inTe
+_20	v'eInte
+_2X	v'eInt
+_3X	t@-*'eInta
+_4X	kwa*'Enta
+_5X	Tinkw'Enta
+_6X	sEs'Enta
+_7X	sEt'Enta
+_8X	otS'Enta
+_9X	nov'Enta
+_0C	T'ientos
+_1C0	T'ien         // exactly one hundred
+_1C	T'iento
+_5C	kinj'Entos
+_7C	s,EtETj'Entos
+_9C	n,ovETj'Entos
+_0M1	m'il
+_1M1	m'il          // no '1' before thousand
+_0M2	mil^'ones
+_1M2	'unmil^'on
+_0M4	_bil^'ones
+_1M4	'unbil^'on
+_0and	i
+_dpt	koma
 
 
 
 
 
 // possessives
-mi	$u+
+mi	mi $u+
 mis	$u+
 tu      $u+
 tus     $u+

dictsource/es_rules

-// translation rules
+// translation rules for Spanish
 // This file is UTF-8 encoded
 
+// Conditional rules
+// ?1  Castilian
+// ?2  Latin America
+
 
 .group a
        _) a (_       a
           a          a
           ai         aI
-          ay         aI
+          ay (K      aI
           ay (_      'aI
 
 
           e          E
           e (_       e
           ei         eI
-          ey         eI
+          ey (K      eI
           ey (_      'eI
        _) eu         eU
 
 
 
 .group v
-       _) v (_       ve
+       _) v (_       uve
           v          v
        _) v          v
        m) v          v
 
 
 .group w
-       _) w (_       d'oBle,u
+       _) w (_       ,uve||d'oBle
           w          w
 
 

dictsource/hr_list

 _1M4  b'ilijUn
 _dpt  _:z'a*Ez_
 
-// Roman numerals
-ii	$abbrev
-iii	$abbrev
-iv	$abbrev
-vii	$abbrev
-viii	$abbrev
-ix	$abbrev
-xi	$abbrev
-xii	$abbrev
-xiii	$abbrev
-xiv	$abbrev
-xvi	$abbrev
-xvii	$abbrev
-xviii	$abbrev
-xix	$abbrev
-xxi	$abbrev
-xxii	$abbrev
-xxiii	$abbrev
-xxiv	$abbrev
-xxvi	$abbrev
-xxvii	$abbrev
-xxviii	$abbrev
-xxix	$abbrev
-xxxi	$abbrev
-xxxii	$abbrev
-xxxiii	$abbrev
-xxxiv	$abbrev
-xxxvi	$abbrev
-xxxvii	$abbrev
-xxxviii  $abbrev
-xxxix	$abbrev
+
 
 // Abbreviations
 aaa	$abbrev

dictsource/hu_list

 ú.n	u:JnEvEzEt:   $dot
 vö	v'EZd||_'Yss2E
 
-?1 i	EJ	$abbrev
-ii	kEttY:  $abbrev
-iii	ha:R2om  $abbrev
-iv	ne:J  $abbrev
-vi	hAt  $abbrev
-vii	he:t  $abbrev
-viii	n^olts  $abbrev
-ix	kilEnts  $abbrev
-xi	tizEnEJ  $abbrev
-xii	tizEnkEttY:  $abbrev
-xiii	tizEnha:Rom  $abbrev
-xiv	tizEnne:J  $abbrev
-xv	tizEnYt  $abbrev
-xvi	tizEnhAt  $abbrev
-xvii	tizEnhe:t  $abbrev
-xviii	tizEnn^olts  $abbrev
-xix	tizEnkilEnts  $abbrev
-xx	hu:s  $abbrev
-xxi	husonEJ  $abbrev
-xxii	husonkEttY:  $abbrev
-xxiii	husonha:Rom  $abbrev
-xxiv	husonne:J  $abbrev
-xxv	husonYt  $abbrev
-xxvi	husonhAt  $abbrev	
-xxvii	husonhe:t  $abbrev
-xxviii	husonn^olts  $abbrev
-xxix	husonkilEnts  $abbrev
-xxx	hARmints  $abbrev
+?1 i	EJ	$abbrev     // option for Roman number
 
 
 // Articles
 nem       $alt2 $strend
 ne        $u+
 
+e         $u          // "-e" question
+e         $atend      // stressed at end of sentence
 
 // Unstressed adverbs prepositions etc.
 néhány    $u+
 on	$u	// on
 en	$u
 ön	$u
+an	$u	// so many persons
+ón	$u
 szor	$u	// times
 szer	$u
 ször	$u
 minden	$u	// all
 részben	$u	// partly
 
+egyik	$u	// one of
+másik	$u	// the other
+
 
 // word pairs
 
 // Main Exceptions List
 //=====================
 
+nato	na:to:
+
 
 

dictsource/hu_rules

         a        A
      _) a (_     %A
 
-   !_n) ato      a:to:      // NATO
-
 .group á
         á        a:
 
 .group b
         b        b
-        bb       b:
+//        bb       b:
 
         biz (A   b'iz,       // bizalmas  etc.
 	biedermeier  bi:dER2ma:jER2
 
 .group d
         d        d
-        dd       d:
+//        dd       d:
         dz       dz
      A) dz (A    d:z
         dzs      dZ
 
 .group e
         e        E
-
   D_-_) es (_    %ES
 
 .group é
 
 .group g
         g        g
-        gg       g:
+//        gg       g:
         gy       J
         ggy      J:
 
      A) gyj (A   J:
 
     _e) gy (es   J:
-    _e) gy (et   J:
+    _e) gy (et_  J:
+    _e) gy (etlen  J:
     _e) gy (ik   J:
     _e) gy (üt   J:
 _minde) gy (ik   J:
 .group k
         k        k
         kk       k:
+        kk (C    kk
      _) köz      kYz
 
 .group l
 .group p
         p        p
         pp       p:
+        pp (C    pp
         p (v     p_
 
      _) potsdam  potsda:m
         r        R2
         rr       RR
         rm       R2_m
+        r (cc    R
 
      _) richa (rd R2iCa:
      _) richá (rd R2iCa:
      -) t (_     t
         t        t
         tt       t:
+        tt (C    tt
         ty       c
 //     s) ty       c:    //
 //     z) ty       c:    // keztyű

dictsource/it_list

 ti	$u+
 si      $u+
 ci	$u+
-vi	$u+
-li	$u+
+vi	vi $u+
+li	li $u+
 
 
 // words that double the following consonant

dictsource/la_list

 sunto $u
 
 
-// numerals 1-100
+// numerals
+
+ii	II	// word, not numeric
+vi	wI	// word, not numeric
 
 i		'u:nUs $capital 
 ii		d'UO $capital 
-iii		t@-*,e:s 
-iv		kw'at:UOR 
-v		kw'i:NkwE 
 vi		s,Eks $capital 
-vii		s'EptEm 
-viii		'Okto: 
-ix		n'OwEm 
-x		d'EkEm 
-
-xi		'u:ndEkIm
-xii		dU'OdEkIm
-xiii		t@-*,e:d'EkIm
-xiv		kwat:w'ORdEkIm
-xv		kw,i:nd'EkIm
-xvi		s'EdEkIm
-xvii		sEpt'EndEkIm
-xviii		dUOdEwIg'IntI
-xix		u:ndEwIg'IntI 
-xx		wi:g'Inti:
-
-xxi		wi:g'Inti:||'u:nUs
-xxii		wi:g'Inti:||d'UO
-xxiii		wi:g'Inti:||t@-*,e:s
-xxiv		wi:g'Inti:||kw'at:UOR
-xxv		wi:g'Inti:||kw'i:NkwE
-xxvi		wi:g'Inti:||s,Eks
-xxvii		wi:g'Inti:||s'EptEm
-xxviii	wi:g'Inti:||'Okto:
-xxix		wi:g'Inti:||n'OwEm
-xxx		t@-*i:g'Inta:
-
-xxxi		t@-*i:g'Inta:||'u:nUs
-xxxii		t@-*i:g'Inta:||d'UO
-xxxiii	t@-*i:g'Inta:||t@-*,e:s
-xxxiv		t@-*i:g'Inta:||kw'at:UOR
-xxxv		t@-*i:g'Inta:||kw'i:NkwE
-xxxvi		t@-*i:g'Inta:||s,Eks
-xxxvii	t@-*i:g'Inta:||s'EptEm
-xxxviii	t@-*i:g'Inta:||'Okto:
-xxxix		t@-*i:g'Inta:||n'OwEm
-xl		kwad@-*a:g'Inta:
-
-xli		kwad@-*a:g'Inta:||'u:nUs
-xlii		kwad@-*a:g'Inta:||d'UO
-xliii		kwad@-*a:g'Inta:||t@-*,e:s
-xliv		kwad@-*a:g'Inta:||kw'at:UOR
-xlv		kwad@-*a:g'Inta:||kw'i:NkwE
-xlvi		kwad@-*a:g'Inta:||s,Eks
-xlvii		kwad@-*a:g'Inta:||s'EptEm
-xlviii	kwad@-*a:g'Inta:||'Okto:
-xlix		kwad@-*a:g'Inta:||n'OwEm
-l		kwi:Nkwa:g'Inta:
-
-li		kwi:Nkwa:g'Inta:||'u:nUs $capital
-lii		kwi:Nkwa:g'Inta:||d'UO
-liii		kwi:Nkwa:g'Inta:||t@-*,e:s
-liv		kwi:Nkwa:g'Inta:||kw'at:UOR
-lv		kwi:Nkwa:g'Inta:||kw'i:NkwE
-lvi		kwi:Nkwa:g'Inta:||s,Eks
-lvii		kwi:Nkwa:g'Inta:||s'EptEm
-lviii		kwi:Nkwa:g'Inta:||'Okto:
-lix		kwi:Nkwa:g'Inta:||n'OwEm
-lx		sEksa:g'Inta:
-
-lxi		sEksa:g'Inta:||'u:nUs
-lxii		sEksa:g'Inta:||d'UO
-lxiii		sEksa:g'Inta:||t@-*,e:s
-lxiv		sEksa:g'Inta:||kw'at:UOR
-lxv		sEksa:g'Inta:||kw'i:NkwE
-lxvi		sEksa:g'Inta:||s,Eks
-lxvii		sEksa:g'Inta:||s'EptEm
-lxviii	sEksa:g'Inta:||'Okto:
-lxix		sEksa:g'Inta:||n'OwEm
-lxx		sEptUa:g'Inta: $capital
-
-lxxi		sEptUa:g'Inta:||'u:nUs
-lxxii		sEptUa:g'Inta:||d'UO
-lxxiii	sEptUa:g'Inta:||t@-*,e:s
-lxxiv		sEptUa:g'Inta:||kw'at:UOR
-lxxv		sEptUa:g'Inta:||kw'i:NkwE
-lxxvi		sEptUa:g'Inta:||s,Eks
-lxxvii	sEptUa:g'Inta:||s'EptEm
-lxxviii	sEptUa:g'Inta:||'Okto:
-lxxix		sEptUa:g'Inta:||n'OwEm
-lxxx		Okto:g'Inta:
-
-lxxxi		Okto:g'Inta:||'u:nUs
-lxxxii	Okto:g'Inta:||d'UO
-lxxxiii	Okto:g'Inta:||t@-*,e:s
-lxxxiv	Okto:g'Inta:||kw'at:UOR
-lxxxv		Okto:g'Inta:||kw'i:NkwE
-lxxxvi	Okto:g'Inta:||s,Eks
-lxxxvii	Okto:g'Inta:||s'EptEm
-lxxxviii	Okto:g'Inta:||'Okto:
-lxxxix	Okto:g'Inta:||n'OwEm
-xc		no:na:g'Inta:
-
-xci		no:na:g'Inta:||'u:nUs
-xcii		no:na:g'Inta:||d'UO
-xciii		no:na:g'Inta:||t@-*,e:s
-xciv		no:na:g'Inta:||kw'at:UOR
-xcv		no:na:g'Inta:||kw'i:NkwE
-xcvi		no:na:g'Inta:||s,Eks
-xcvii		no:na:g'Inta:||s'EptEm
-xcviii	no:na:g'Inta:||'Okto:
-xcix		no:na:g'Inta:||n'OwEm
-c		k'EntUm
-
-// This list could be expanded further, however programming
-// the pronunciation would be more efficient.
 
+_0	n'Ulla
+_1	'u:nUs 
+_2	d'UO 
+_3	t@-*,e:s 
+_4	kw'at:UOR 
+_5	kw'i:NkwE 
+_6	s,Eks
+_7	s'EptEm 
+_8	'Okto: 
+_9	n'OwEm 
+_10	d'EkEm 
+
+_11	'u:ndEkIm
+_12	dU'OdEkIm
+_13	t@-*,e:d'EkIm
+_14	kwat:w'ORdEkIm
+_15	kw,i:nd'EkIm
+_16	s'EdEkIm
+_17	sEpt'EndEkIm
+_18	dUOdEwIg'IntI
+_19	u:ndEwIg'IntI 
+
+_2X	wi:g'Inti:
+_3X	t@-*i:g'Inta:
+_4X	kwad@-*a:g'Inta:
+_5X	kwi:Nkwa:g'Inta:
+_6X	sEksa:g'Inta:
+_7X	sEptUa:g'Inta:
+_8X	Okto:g'Inta:
+_9X	no:na:g'Inta:
+_0C	k'EntUm
+_0M1	m'IllE

dictsource/pt_list

 _ú	,u&g'udU
 _ü	,utR'em&
 
+i	i        // speak i v x as letters, not Roman numbers
+v	ve
+x	Sis#
+
+
 _cap	maI'uskulU
 _?A	l'etR&
 _??	s'imbolU
 // numeric
 //********
 
-?1_0	z'Eru
-?2_0	z'E*U
-?1_1	'um
-?2_1	'uN
-_2	d'oIs#
+?1 _0	z'Eru
+?2 _0	z'E*U
+?1 _1	'um
+?2 _1	'uN
+   _2	d'oIs#
 ?1 _3	tr'es#
 ?2 _3	tR'es#
 ?1 _4	kw'atru
 ?2 _4	kw'atRu
-_5	s'iNku
-_6	s'eIs#
-?1_7	s'Ety
-?2_7	s'EtSi
-_8	'oItu
-?1_9	n'Ovy
-?2_9	n'Ovi
-10	d'Es#
+   _5	s'iNku
+   _6	s'eIs#
+?1 _7	s'Ety
+?2 _7	s'EtSi
+   _8	'oItu
+?1 _9	n'Ovy
+?2 _9	n'Ovi
+   _10	d'Es#
 
 ?1_11	'oNzy
 ?2_11	'oNzi
 unts      $abbrev
 usb       $abbrev
 
-// roman numerals
-?1 ii	d'oIs#       $abbrev    
-?1 iii	tr'es#      $abbrev
-?1 iv	kw'atru       $abbrev
-
 
-dr                           $dot
-dra                           $dot
-mr                           $dot
-mrs                           $dot
-prof       pr'Of         $dot
-sr                           $dot
-?1 sra                ,Es;'Errj'a           $dot
-?2 sra              $dot
+dr	$dot
+dra	$dot
+mr	$dot
+mrs	$dot
+prof	pr'Of	$dot
+sr	$dot
+?1 sra	,Es;'Errj'a	$dot
+?2 sra	$dot
 
 // articles
 o	$u+

dictsource/pt_rules

           únh        ''u~n^
 
 .group v
-       _) v (_       ve
           v          v
           v (v
 
 
 
 .group x
-       _) x (_       Sis#
           x (A       S
        A) x (A       ks
       _e) x (A       z

dictsource/ro_rules

+
 // translation rules for Romanian
 // This file is UTF-8 encoded
 
 
 
+
 .group a
         a        a
         ai       aI
         au (_    'aU
         a (AAA   a
      &) a (_S1q  a
+
     gi) a (_S1q  'a
-    V@) a (_     'a	
+    V@) a (N_    'a	
     ăi) a (_     'a
      @) abil (_  'abil
      _) anti (@P4  anti
      @) ale (_   'ale
+  @ion) a (_     'a
+
 
 .group ă
         ă        @
-        ăi       @I
+        ăi (_    @I
         ău       @U  // ?? only at end of word ??
-        ău (_    @U
-
-        ăi (m_   @'i
-        ăi (t_   @'i
-        ăi (ţi_  @'i
-
-
-        ăi (te   @'i
-        ăi (n_   @'i
-        ăi (na_  @'i
-        ăi (ni_  @'i
-        ăi (ne_  @'i
+        ău (_    '@U
+     @) ăte (_S3t @te
+     @) ătă (_S3t @t@
+     @) ătul (_S4t @tul
+     @) ătele (_S5t @tele
+     @) ătului (_S6t @tuluI
+     @) ătelor (_S6t @telor
      &) ă (_S1   @
+   &ec) ă (_     @
+   @ăC) ă (_     @
+   @er) ă (_     @
+   @iţ) ă (_     @
+     @) ă (Că_   =@
         ări (A_  @*'i
         ării (_  @*'iI
         ări (AA_ @*'i
-     @) ăt (_S2t     @t
-     @) ătă (_S3t    @t@
-     @) ătul (_S4t   @tul
-     @) ătele (_S5t  @tele
-     @) ătului (_S6t @tuluI
-     @) ătelor (_S6t @telor
+     @) ăie (_   =@Ie
 
 
 .group â
         â        y
         âi       yI
         âu       yU
+     @) âie (_ =yIe
 
 
 .group b
         b        b
-     @) b (i_    b;
+        bb       b
 
+     @) b (i_    b;
      @) b (il_   =b
+     @) b (ila_ =b
+     @) b (ili_ =b
+     @) b (ilii_ =b
+     @) b (ilului_ =b
+     @) b (ililor_ =b
+     @) b (ilile_ =b
     mo) bil (_   b'il
 
 
     @@) col (_   =kol
     @@) coli (_S1t  =I^
     @@) cole (_S1t  e
-
-
      @) cni (_   kn'i
      @) cnea (_  kne'a
 
 
 .group d
         d        d
+        dd       d
      @) d (i_    d;
 
 
 .group e
         e        e
-        ei       eI
-//        eu       eU   // only at end of word ??
-//        eu       e[u    // ??
+     _) ei       eI
+        ei (_ eI
+//      eu       eU   // only at end of word ??
+//      eu       e[u    // ??
         eu (_    'eU
         ea       ea
      &) ea (_S2q ea
      &) eSi      eaj
         eau      eaw
-
         eo       eo
-        eoa      eO'a
-
         e (AAA   e
+     @) eş (_    =eS
+
+     @) eşul (_  =eSul
+     @) eşi (_   =eSI^
+     @) eşului (_ =eSuluI
+     @) eşilor (_ =eSilor
+
+     @) ec (a_   =ek
+     @) ec (ă_   =ek
+     @) ec (ii_  =ek
+     @) ec (ile_ =ek
+     @) ec (ilor_ =ek
+   @At) ec (i@_  etS
+   @At) ec (A_   ek
+   @At) ec (ii_  eTtS 
+//   &) es (_S1q e
     &C) e (re_   %e
+   _ac) ee (a	 eI
+
+     r) ei (er_  'eI
+
      @) e (re_   =e
      @) e (ri_   =e
      @) e (rea_  =e
+     @) e (rii_  =e
      i) e (re_   e
      i) e (ri_   e
+     i) e (rii_  e
      i) e (rea_  e
-
-   _ac) ee (a	 eI
-     r) ei (er_  'eI
-
-     @) eş (_    =eS
-
-     @) eşul (_   =eSul
-     @) eşi (_    =eSI^
-     @) eşului (_ =eSuluI
-     @) eşilor (_ =eSilor
-// &) es (_S1q     e
-
+     @) e (ra_   =e
+     @) e (ră_   =e
+    @f) e (ra_   e
+    @f) e (ra_   e
+     i) e (ră_   e
+     i) e (ra_   e
+        eoa      eO'a 
 
 
 .group f
         f        f
+        ff       f
      @) f (i_    f;
 
 
 .group i
         i        i
         i (A     j
+
+     @) i (e_    j
     &C) i (_N$1q i/
      &) i (_N$1q i
      &) i (-N$1q i        // don't reduce to [I^] if a hyphen follows
-
      ţ) i (-Că   i
      ţ) i (-Ci   i
+
+     ţ) i (-l    i
+     ţ) i (-ne   i 
     &') i (_     i
    ţi-) i (_     _        // combined with previous word as [iI]
     Cr) i (_     i
-    V&) i (_     'i       // verb
+    V&) i (N_    'i       // verb
     ăC) ia (_    'ia
     âC) ia (_    'ia
-
     ăC) ie (_    'ie
     âC) ie (_    'ie
-     @) ie (_S1q  ie
+
+     g) i (ne_   =i
+     g) i (nea_  =i
+     c) i (uC 
+    cţ) ie (_    =ie 
+    cţ) iei (_   =ieI
+    cţ) ia (_    =ja
+  graf) ie (_    'ie
+  graf) ia (_    'ia
+  graf) iei (_   'ieI
+  graf) ii (_    'iI
+  scop) ie (_    'ie
+  scop) ia (_    'ia
+  scop) iei (_   ieI
+  scop) ii (_    'iI
+    er) ia (_    'ia
+    er) iei (_   'ieI
+    on) ie (_    'ie
+    on) ia (_    'ia
+    on) iei (_   'ieI
+    uş) ie (_    'ie
+    uş) ia (_    'ia
+    uş) iei (_   'ieI
+//   @) ie (_S1q ie
 
         iei (_   'ieI
         ii       iI
         iu       ju
 
      &) ic (_    =ik
+     &) ica (_   =ika
      &) ice (_   =itSe
-     &) icile (_  =itSile 
+     &) icile (_ =itSile 
      &) icilor (_ =itSilor
      @) icii (_  =itSiI
 
+   scr) i (e 'i
      &) ing (_   %iNg     // foreign words
 
         i (t_    'i
-        ioa iO'a
+        ioa      iO'a
+
+     @) işte (_  =iSte
+     @) iştea (_ =iStea
+     @) i (ţă_++++++++ =i
+     @) i (ţe_   =i
+     @) i (ţei_  =i
+     @) i (ţelor_ =i
+     -) i (_     j
+     @) i (ţele_ =i
 
 
 .group î
 
 .group l
         l        l
+        ll       l
      @) l (i_    l;
      _) l (_     le
      -) l (_     l         // -l
 .group m
      _) m (_     me
         m        m
+        mm       m
      @) m (i_    m;
      _) m (-a    m
 
      @) n (tic_  =n
      @) n (ic_   =n
      _) n (-a     n
+     -) n (_     n
 
-     -) n (_ n
+@itudi) ne (_S2t ne
+@itudi) nea (_S3t nea
+@itudi) nii (_S3t niI
+@itudi) ni (_S2t ni 
+@itudi) nile (_S4t nile
+@itudi) nilor (_S5t nilor
 
 
 .group o
         o        o
         oa       Oa
         oai      Oaj
-        oi       oI
+     _) oi       oI
         oi (_    'oI
         ou       oU
         ou (_    'oU
         o (AAA   o
 
-     @) oi (t_   o'i
-     @) oi (tă_  o'i
-     @) oi (ţi_  o'i
-     @) oi (te_  o'i
-
 
 .group p
         p        p
+        pp       p
      @) p (i_    p;
 
 
-
 .group q
         q        k
         qu       kw
 
 
 .group r
-        r          r
-     A) r (A       *
-     C) r (A       @-*
-        rr         *r
-     A) r (i_      *;
-
-     _) re (@P2v   re
-     _) re (gul    re
-    @u) ri (_NS2t   rI^
+        r        r
+        rr       r
+     A) r (A     *
+     C) r (A     @-*
+        rr       *r
+     A) r (i_    *;
+
+     _) re (@P2T re
+     _) re (gul  re
+    @u) ri (_NS2t  rI^
+     C) ri (_    @-*i
+    @u) rii (_S3t riI
     @u) rile (_S4t rile
-     @) rilor (_S5t rilor
+    @u) rilor (_S5t rilor
 
-     C) ri (_ @-*i
      @) răm (_S1t m
      @) răţi (_S4t *@tsI^
+    @u) re (_S2t *e
 
 
 .group s
      _) s (_     se
      _) s (-a    s
         s        s
-     @) s (i_    s;
+     -) s (_     s
+        sh       S
+        ss       s
 
      &) sprezece (_S8t sp@-*ez,etSe
 
      @) sem (_S1t   m 
      @) seră (_S2t  *@
-     @) serăm (_S3t *@m 
+     @) serăm (_S3t *@m
      @) serăţi (_S4t *@tsI^
      @) seşi (_S2t   SI^
-     @) sni (_   sn'i
+     @) sni (_ sn'i
 
 
 .group ş
         tt (_    t
      @) t (i_    t;
 
+        ts       ts
+        tz       ts
+        tt       t
+
 
 .group ţ
         ţ        ts
 .group u
         u        u
         u (A     w
-        ui       uI
+     _) ui       uI
+        ui (_ uI
         ui (e    wi
-    V&) ui (_    u'i
-        ui (m_   u'i
-        ui (t_   u'i
-        ui (ţi_  u'i
-        ui (se_  u'i
-     &) ul (_S2t ul
-
-   @ic) ul (_S2  ul
-   @ic) uri (_S3 urI^
+    V&) ui (_N    u'i
+
+     &) ul  (_S2t ul
+   @ic) ul (_S2t  ul
+   @ic) ule (_S3t ule
+   @ic) ului (_S4t uluI
+   @ic) uri (_S3  urI^
    @ic) urile (_S5 urile
 
+     &) ului (_S4t uluI
+     @) us (_S2t us
+     @) usul (_S4t usul
+     @) usuri (_S5t usuri
+     @) usului (_S6t usuluI
+     @) usurile (_SS´7t usurile
+     @) usurilor (_S8t usurilor
+   @ăC) u (N_    'u
+     @) um (_S2t um
+     @) umul (_S4t umul
+     @) umului (_S6t umuluI
+     @) umuri (_S5t umuri
+     @) umurile (_S7t umurile
+     @) umurilor (_S8t umurilor
+     c) um (_    um
+     @) uie (_   =wie
 
 
 .group v
         x        ks
      A) x (A     gz
     ma) x (i     ks
-     e) x (e     ks
   cone) x (iu    ks
      i) x (e     ks
      o) x (e     ks       // boxele
+     u) x (a     ks
+     a) x (a     ks
+     a) x (ă     ks
+     a) x (e     ks
+     i) x (a     ks
+     o) x (a     ks
+     A) x (â     ks
+     @) x (ibil  ks
      o) x (i     ks       // oxigen
      a) x (ia    ks       // axial
    ale) x        ks
 
 .group z
         z        z
+        zz       z
      @) z (i_    z;
 
 
-
 .group
         ö        Y
-
-        %        protSent
     D_) %        la_sut@
+        %        p@-*otSent
+        $        dolar
+        +        plus
+        &        ampersand
+        ©        k'apiraIt 
+        @        a*'ond
+        /        sleS
+        €        'eU*o
+        °        g@-*ade
 
         , (D     v'irgul@
 
     *)  * (*
         **
    **)  * (_     _::
+
+

phsource/mbrola/ro1

 0  r/ NULL   0   r
 0  r- NULL   0   r
 0  g- NULL   0   NULL
-0  w/ NULL   0   w
+0  w NULL   0   u
+0  w/ NULL   0   u
 0  *  NULL   0   r
 0  ;  NULL   0   NULL
 0  a  e      15  a  _
 0  a  NULL   0   a
 0  @- NULL   0   NULL
+0  j/ NULL   0   j
 0  I^ _      0   I
 0  I^ NULL   20  I  _
 0  @U NULL   60  @  w
 0  aU NULL   60  a  w
-0  eU NULL   60  e  w
+0  eU NULL   80  e  u
 0  iU NULL   60  i  w
-0  yU NULL   60  l  w
+0  yU NULL   60  1  w
 0  oU NULL   60  o  w
 0  @I NULL   60  @  j
 0  iI NULL   60  i  j
 0  aI NULL   60  a  j
-0  eI NULL   60  e  j
+0  eI NULL   100  e i
 0  oI NULL   60  o  j
 0  uI NULL   60  u  j
 0  yI NULL   60  1  j

phsource/ph_english_us

   vowel  starttype (@)  endtype (@)
   length 230
   formants vwl_en_us/3_us
-  reduceto @  0
+//  reduceto @  0
   linkout r-
 endphoneme
 

src/numbers.cpp

 	char capital[20];
 	char ph_buf[30];
 	char ph_buf2[50];
-	static char single_letter[8] = {0};
+	static char single_letter[10] = {0,0};
 
 	ph_buf[0] = 0;
 	capital[0] = 0;
 
 	n_bytes = utf8_in(&letter,word,0);
 
+	if((letter & 0xfff00) == 0x0e000)
+	{
+		letter &= 0xff;   // uncode private usage area
+	}
+
 	if(control > 2)
 	{
 		// include CAPITAL information
 	}
 	letter = towlower(letter);
 
+	if((letter <= 32) || iswspace(letter))
+	{
+		// lookup space as _&32 etc.
+		sprintf(&single_letter[1],"_#%d ",letter);
+		Lookup(&single_letter[1],ph_buf);
+		strcat(phonemes,ph_buf);
+		return(n_bytes);
+	}
+
 	len = utf8_out(letter,&single_letter[2]);
-	single_letter[2+len] = ' ';
+	single_letter[len+2] = ' ';
 
 	next = RULE_SPELLING;
 	if(word[n_bytes] == ' ')
 
 
 
+int Translator::TranslateRoman(char *word, char *ph_out)
+{//=====================================================
+	int c;
+	char *p;
+	int acc;
+	int prev;
+	int value;
+	int subtract;
+	unsigned int flags;
+	char number_chars[N_WORD_BYTES];
+
+	static char *roman_numbers = "ixcmvld";
+	static int roman_values[] = {1,10,100,1000,5,50,500};
+ 
+	if((langopts.numbers & NUM_ROMAN) == 0)
+		return(0);
+
+	acc = 0;
+	prev = 0;
+	subtract = 0x7fff;
+
+	while((c = *word++) != ' ')
+	{
+		if((p = strchr(roman_numbers,c)) == NULL)
+			return(0);
+
+		value = roman_values[p - roman_numbers];
+
+		if((prev==5) || (prev==50) || (prev==500))
+		{
+			if(value >= prev)
+				return(0);
+		}
+		if((prev != 0) && (prev < value))
+		{
+			if(((acc % 10) != 0) || ((prev*10) < value))
+				return(0);
+			subtract = prev;
+			value -= subtract;
+		}
+		else
+		if(value >= subtract)
+			return(0);
+		else
+			acc += prev;
+		prev = value;
+	}
+	acc += prev;
+	if(acc < 2)
+		return(0);
+
+	if(acc > langopts.max_roman)
+		return(0);
+
+	Lookup("_roman",ph_out);   // precede by "roman" if _rom is defined in *_list
+	p = &ph_out[strlen(ph_out)];
+
+	sprintf(number_chars," %d ",acc);
+	TranslateNumber(&number_chars[1],p,&flags,0);
+	return(1);
+}  // end of TranslateRoman
 
 
 int Translator::LookupNum2(int value, int control, char *ph_out)
 		{
 			if((thousandplex > 0) && (value < 1000))
 			{
-				if(langopts.numbers & 0x40000)
+				if(langopts.numbers2 & 0x100)
 				{
 					if((thousandplex == 1) && (value >= 100))
 					{

src/readclause.cpp

 						c1 = j;
 					}
 				}
+				if((sayas_mode == 0x14) && (c1 <= 0x20))
+				{
+					c1 += 0xe000;  // move into unicode private usage area
+				}
 			}
 			else
 			if(c1 == '<')

src/synthdata.cpp

 #include "translate.h"
 #include "wave.h"
 
-const char *version_string = "1.27.05  05.Jul.07";
+const char *version_string = "1.27.06  07.Jul.07";
 const int version_phdata  = 0x012701;
 
 int option_device_number = -1;

src/tr_english.cpp

 	langopts.stress_rule = 0;
 	langopts.word_gap = 0;
 
-	langopts.numbers = 0x41;
-
+	langopts.numbers = 0x41 + NUM_ROMAN;
 }
 
 

src/tr_languages.cpp

 			tr->langopts.param[LOPT_PREFIXES] = 1;
 			memcpy(tr->stress_lengths,stress_lengths_de,sizeof(tr->stress_lengths));
 		
-			tr->langopts.numbers = 0x11c19;
+			tr->langopts.numbers = 0x11c19 + NUM_ROMAN;
 			SetLetterVowel(tr,'y');
 		}
 		break;
 			tr->langopts.unstressed_wd2 = 2;
 			tr->langopts.param[LOPT_SONORANT_MIN] = 120;  // limit the shortening of sonorants before short vowels
 
-			tr->langopts.numbers = 0x529;
+			tr->langopts.numbers = 0x529 + NUM_ROMAN;
 		}
 		break;
 
 
 			tr->langopts.stress_rule = 6;      // stress on last heaviest syllable
 			tr->langopts.stress_flags =  0x10004;   // use 'diminished' for unstressed final syllable
-			tr->langopts.numbers = 0x811 + 0x40000;
+			tr->langopts.numbers = 0x811;
+			tr->langopts.numbers2 = 0x100;
 			tr->letter_bits_offset = OFFSET_DEVANAGARI;
 			tr->langopts.replace_chars = replace_chars_hi;
 			tr->langopts.replacement_chars = replacement_chars_hi;
  			tr->langopts.max_initial_consonants = 5;
 			tr->langopts.spelling_stress = 1;
 
-			tr->langopts.numbers = 0x1c0d + 0x4000;
+			tr->langopts.numbers = 0x1c0d + 0x4000 + NUM_ROMAN;
 			tr->langopts.numbers2 = 0x4a;  // variant numbers before thousands,milliards
 			tr->langopts.replace_chars = replace_cyrillic;
 			tr->langopts.replacement_chars = replace_cyrillic_latin;
 			tr->langopts.param[LOPT_COMBINE_WORDS] = 99;  // combine some prepositions with the following word
 			tr->langopts.long_stop = 130;
 
-			tr->langopts.numbers = 0x1809;
+			tr->langopts.numbers = 0x1809 + NUM_ROMAN;
 			SetLetterVowel(tr,'y');
 			tr->langopts.spelling_stress = 1;
 //SetLengthMods(tr,3);  // all equal
 			tr->langopts.param[LOPT_IT_DOUBLING] = 2;    // double the first consonant if the previous word ends in a stressed vowel
 			tr->langopts.param[LOPT_SONORANT_MIN] = 130;  // limit the shortening of sonorants before short vowels
 			tr->langopts.param[LOPT_REDUCE] = 1;        // reduce vowels even if phonemes are specified in it_list
-			tr->langopts.numbers = 0x2709 + 0x800;
+			tr->langopts.numbers = 0x2709 + 0x800 + NUM_ROMAN;
 		}
 		break;
 
 			tr->langopts.unstressed_wd1 = 0;
 			tr->langopts.unstressed_wd2 = 2;
 			tr->langopts.param[LOPT_DIERESES] = 1;
+			tr->langopts.numbers = 0x1 + NUM_ROMAN;
+			tr->langopts.max_roman = 5000;
 		}
 		break;
 
 //			tr->langopts.vowel_pause = 1;
 			tr->langopts.stress_rule = 3;        // stress on final syllable
 			tr->langopts.stress_flags =  0x6 | 0x10 | 0x20000; 
-			tr->langopts.numbers = 0xa69 + 0x2000;
+			tr->langopts.numbers = 0xa69 + 0x2000 + NUM_ROMAN;
 			tr->punct_to_tone[0][3] = 2;   // use exclamation intonation
 			SetLetterVowel(tr,'y');
 		}
 			tr->charset_a0 = charsets[2];   // ISO-8859-2
 			tr->langopts.replace_chars = replace_chars_ro;
 			tr->langopts.replacement_chars = replacement_chars_ro;
-			tr->langopts.numbers = 0x1829+0x6000;
+			tr->langopts.numbers = 0x1829+0x6000 + NUM_ROMAN;
 			tr->langopts.numbers2 = 0x1e;  // variant numbers before all thousandplex
 		}
 		break;
 			tr->langopts.stress_rule = 2;
 			tr->langopts.stress_flags =  0x6 | 0x10; 
 
-			tr->langopts.numbers = 0x408e1;
+			tr->langopts.numbers = 0x8e1;
+			tr->langopts.numbers2 = 0x100;
 		}
 		break;
 
 	langopts.param[LOPT_PREFIXES] = 1;
 	SetLetterVowel(this,'y');  // add 'y' to vowels
 
-	langopts.numbers = 0x0d1;
+	langopts.numbers = 0x0d1 + NUM_ROMAN;
 	memcpy(stress_lengths,stress_lengths2,sizeof(stress_lengths));
 }
 

src/translate.cpp

 	langopts.length_mods0 = length_mods_en0;
 	langopts.long_stop = 100;
 
+	langopts.max_roman = 49;
 	langopts.thousands_sep = ',';
 	langopts.decimal_sep = '.';
 
 			found = TranslateNumber(word,phonemes,&dictionary_flags,wflags);
 		}
 
-		if((wflags & FLAG_ALL_UPPER) && (clause_upper_count <= clause_lower_count) &&
+		if(!found & ((word_flags & FLAG_UPPERS) != FLAG_FIRST_UPPER))
+		{
+			// either all upper or all lower case
+			found = TranslateRoman(word,phonemes);
+		}
+
+		if(!found && (wflags & FLAG_ALL_UPPER) && (clause_upper_count <= clause_lower_count) &&
 			!(dictionary_flags & (FLAG_ABBREV | FLAG_SKIPWORDS)) && (word_length>1) && (word_length<4) && iswalpha(first_char))
 		{
 			// An upper case word in a lower case clause. This could be an abbreviation.
 	int phoneme_mode = 0;
 	int dict_flags;        // returned from dictionary lookup
 	int word_flags;        // set here
+	int next_word_flags;
 	int embedded_count = 0;
 	int letter_count = 0;
 	char *word;
 	word_count = 0;
 	single_quoted = 0;
 	word_flags = 0;
+	next_word_flags = 0;
 	expect_verb=0;
 	expect_past=0;
 	expect_verb_s=0;
 			}
 		}
 		else
-		if((option_sayas2 & 0xf0) != 1)
+		if((option_sayas2 & 0xf0) != 0x10)
 		{
 if((c == '/') && (langopts.testing & 2) && isdigit(next_in) && IsAlpha(prev_out))
 {
 						c = ' ';      // change from upper to lower case, start new word at the last uppercase
 						prev_in2 = c;
 						source_index = prev_source_index;  // unget
+						next_word_flags |= FLAG_NOSPACE;
 					}
 //#endif
 				}
 				for(k=j; charix[k]!=0; k++);
 				words[word_count].length = k-j;
 
-				word_flags = 0;
+				word_flags = next_word_flags;
+				next_word_flags = 0;
 				pre_pause = 0;
 				word_mark = 0;
 				all_upper_case = FLAG_ALL_UPPER;

src/translate.h

 // wordflags, flags in source word
 #define FLAG_ALL_UPPER     0x1    /* no lower case letters in the word */
 #define FLAG_FIRST_UPPER   0x2    /* first letter is upper case */
+#define FLAG_UPPERS        0x3    // FLAG_ALL_UPPER | FLAG_FIRST_UPPER
 #define FLAG_HAS_PLURAL    0x4    /* upper-case word with s or 's lower-case ending */
 #define FLAG_PHONEMES      0x8    /* word is phonemes */
 #define FLAG_LAST_WORD     0x10   /* last word in clause */
 #define FLAG_STRESSED_WORD 0x20   /* this word has explicit stress */
 #define FLAG_EMBEDDED      0x40   /* word is preceded by embedded commands */
 #define FLAG_HYPHEN        0x80
+#define FLAG_NOSPACE       0x100  // word is not seperated from previous word by a space
 #define FLAG_DONT_SWITCH_TRANSLATOR  0x1000
 #define FLAG_SUFFIX_REMOVED  0x2000
 #define FLAG_HYPHEN_AFTER    0x4000
 	unsigned char *length_mods;
 	unsigned char *length_mods0;
 
+#define NUM_ROMAN   0x20000
 	// bits0-1=which numbers routine to use.
 	// bit2=  thousands separator must be space
 	// bit3=  , decimal separator, not .
 	// bit12=allow space as thousands separator (in addition to langopts.thousands_sep)
 	// bits13-15  post-decimal-digits 0=single digits, 1=(LANG=it) 2=(LANG=pl) 3=(LANG=ro)
 	// bit16=dot after number indicates ordinal
-	// bit18=special word for 100,000s LANG=sw
+	// bit17=recognize roman numbers
 
 	int numbers;
 
 	// bits 1-4  use variant form of numbers before thousands,millions,etc.
 	// bit6=(LANG=pl) two forms of plural, M or MA
 	// bit7=(LANG-ru) use MB for 1 thousand, million, etc
+	// bit8=(LANG=sw) special word for 100,000s
 	int numbers2;
-
+	int max_roman;
 	int thousands_sep;
 	int decimal_sep;
 	int intonation;          // 1=tone language
 	int LookupNum3(int value, char *ph_out, int suppress_null, int thousandplex, int prev_thousands);
 	int LookupThousands(int value, int thousandplex, char *ph_out);
    int TranslateNumber_1(char *word1, char *ph_out, unsigned int *flags, int wflags);
+	int TranslateRoman(char *word, char *ph_out);
 
 	void InitGroups(void);
 	void AppendPhonemes(char *string, int size, const char *ph);