mahta.fetrat
/
HomoFast-eSpeak-Persian
Watch 1
Star 0
Fork 0
Code Issues 0 Pull Requests 0 Releases 0 Wiki Activity
Browse Source

[1.39.26] 

Added languages: ca (Catalan) lv (Latvian)
Numbers: allow for languages which have numbers for 100,000 and (or not) 10,000,000 (eg. hi, sw)
Phoneme [ts] new files in phsource/ustop
*_rules: fix crash for lettergroups numbers greater than 10, and better error reporting.
Languages: eo, jbo: adjust syllable lengths.
Language: en, fix "1st, 2nd, 3rd, 5th".  But a full ordinal number feature is still needed.


git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@202 d46cf337-b52f-0410-862d-fd96e6ae7743
master
jonsd 17 years ago
parent
084cd2c17d
commit
a7501374b2
46 changed files with 699 additions and 195 deletions
Split View Show Diff Stats
  1. 34
    10
      dictsource/dict_phonemes
  2. 24
    16
      dictsource/en_list
  3. 0
    2
      dictsource/en_rules
  4. 23
    1
      dictsource/eo_list
  5. 6
    9
      dictsource/eo_rules
  6. 1
    1
      dictsource/es_list
  7. 5
    5
      dictsource/hi_list
  8. 9
    9
      dictsource/jbo_list
  9. 20
    17
      dictsource/jbo_rules
  10. 11
    10
      dictsource/sw_list
  11. 4
    0
      espeak-data/voices/ca
  12. 6
    0
      espeak-data/voices/lv
  13. 73
    46
      phsource/compile_report
  14. 1
    0
      phsource/ph_base2
  15. 67
    0
      phsource/ph_catalan
  16. 8
    7
      phsource/ph_esperanto
  17. 1
    0
      phsource/ph_german
  18. 1
    0
      phsource/ph_hungarian
  19. 217
    0
      phsource/ph_latvian
  20. 1
    1
      phsource/ph_lojban
  21. 1
    0
      phsource/ph_polish
  22. 1
    0
      phsource/ph_romanian
  23. 1
    0
      phsource/ph_russian
  24. 1
    1
      phsource/ph_tamil
  25. 10
    4
      phsource/phonemes
  26. BIN
      phsource/ustop/ts2.wav
  27. BIN
      phsource/ustop/ts_.wav
  28. BIN
      phsource/vdiph/aai
  29. BIN
      phsource/vdiph/aau
  30. BIN
      phsource/vdiph/eei_2
  31. BIN
      phsource/vdiph/vowelchart.png
  32. BIN
      phsource/vowel/aa
  33. BIN
      phsource/vowel/i_3
  34. BIN
      phsource/vowel/i_6
  35. BIN
      phsource/w/w_
  36. 2
    1
      src/compiledata.cpp
  37. 25
    9
      src/compiledict.cpp
  38. 13
    1
      src/dictionary.cpp
  39. 12
    20
      src/numbers.cpp
  40. 3
    2
      src/phoneme.h
  41. 5
    1
      src/phonemelist.cpp
  42. 5
    2
      src/setlengths.cpp
  43. 1
    1
      src/synthdata.cpp
  44. 35
    8
      src/tr_languages.cpp
  45. 62
    6
      src/translate.cpp
  46. 10
    5
      src/translate.h

+ 34
- 10
dictsource/dict_phonemes View File

@@ -107,7 +107,7 @@ x z Z
Dictionary eo_dict

@- a aI aU e eI eU i
o OI u uI
o oI u uI

* b d dZ f g h j
k l m n p r R s
@@ -237,12 +237,13 @@ ts v w w2 z

Dictionary jbo_dict

@ a aI aU e eI i l-
o OI r- u
@ @- a aI aU e eI i
l- o oI r- u

b d dZ f g h j k
l m n N p R s S
t tS v w x z Z
* - b d dZ f g h
j k l m n N p R
s S t tS v w x z
Z


Dictionary ko_dict
@@ -515,8 +516,8 @@ tS v w z
Dictionary lv_dict

a a: ai au e e: ei i
i: ie iu o o: oi u u:
ui uo
i: ie iu o o: u u: ui
uo

: b c d dZ f g h
j J k l l^ m n n^
@@ -543,5 +544,28 @@ y
: b c ch d d. dZ f
g h j J k kh l l.
m n N n. n^ p ph R
R2 s s. S; t t. t.h th
tS v w z
R2 s S s. S; t t. t.h
th tS v w z


Dictionary om_dict

@# a a: E e I I# i:
O o: U u:

** : a# b B c` d d`
dZ f g h j k k` l
m n n^ p p` R s S
t t` tS v w z


Dictionary ca_dict

@ a a# aI e E i o
O u U

* ** : b B d D dZ;
f g j J J^ k l l^
m n N n^ p Q r R
R2 s S; t T tS ts tS;
v v# w z Z Z;

+ 24
- 16
dictsource/en_list View File

@@ -108,22 +108,23 @@ _% p3s'Ent
?5 _% pVRs'Ent
_& amp@sand
_' kwoUt
_( lEftbrakI2t
_) raItbrakI2t
_[ lEftbrakI2t
_] raItbrakI2t
_( lEftpa2rEn
_) raItpa2rEn
_* ast@rIsk
_+ plVs
_, k0m@
_- h,aIf@n
?3 _- daS
_. d0t
_: koUl@n
_; sEmIk'oUl@n
_< ElaNg@L
_< lEsDan
_= i:kw@Lz
_> A@aNg@L
_> greIt@Dan
_? kwEstS@n
_@ at|saIn
_[ lEftskwe@
_] raItskwe@
_^ s3:kVmfl,Eks
?5 _^ sIRkVmfl,Eks
__ 'Vnd3sk,o@
@@ -183,7 +184,8 @@ _₠ jU@roU
₨ ru:pi:
₩ w0n
§ sEkS@n
¶ par@graf // in en_rules
¶ par@graf
¶¶ par@grafs
¤ kVr@nsIsaIn
© k0pIraIt
® rEdZIst3d
@@ -280,6 +282,11 @@ _0M3 b'Ili@n
_dpt pOInt
_roman roUm@n_

(1 st) f3:st
(2 nd) sEk@nd
(3 rd) T3:d
(5 th) fIT


// ABBREVIATIONS
//**************
@@ -311,6 +318,7 @@ ie aIi:_! $pause $only
i.e aIi:_! $pause
irc $abbrev
lbs paUndz
LBS $abbrev
ltd lImI2tI2d
mc m@k $alt2 // combine with the following word
oem $abbrev
@@ -336,19 +344,19 @@ xxx $abbrev // not roman 30
xy $abbrev


nd $only
rd $only
th $only
st $only
nd $only
rd $only
th $only
st $only

mr mIst3 $dot
mrs mIsI2z $dot
//ms mIz $dot
dr d0kt3 $dot $capital
lt $dot
prof $dot
rev $dot
st $dot
dr $dot
lt $dot
prof $dot
rev $dot
st $dot

// not abbreviations when all-caps
all O:l $allcaps

+ 0
- 2
dictsource/en_rules View File

@@ -5368,5 +5368,3 @@
%) % (%
%%) % (_ _::

// extra symbols (mostly in en_list)
¶¶ par@grafs

+ 23
- 1
dictsource/eo_list View File

@@ -29,7 +29,28 @@ _0M3 mili'ardo
_0M4 duili'ono
_dpt komo_

// accent names
_acu dekst@-*akoRno
_brv hoko
_cir tsiRkumflekso
_dia dupunktasupeRsigno
_grv maldekst@-*akoRno
_tld tildo


_?? liteRo
_. punkto
_: dupunkto
_; punktokomo
_, komo
_? demandosigno
_- dividost@-*eko
~ tildo
° gradoI
% p@-*otsento
@ po
& kaI


v vo // not Roman numbers
x ikso
@@ -100,7 +121,8 @@ tra $u $pause

// misc adverbs etc
eĉ $brk
(pli ol) pliol
pli pl,i
plej pl,eI

// pronouns
mi $u

+ 6
- 9
dictsource/eo_rules View File

@@ -102,7 +102,9 @@
.group o
_) o(_ o
o o
oj OI
oj oI
-) oj (_ %oI
-) on (_ %on
.group p
_) p(_ po
@@ -203,18 +205,13 @@
\.) .


. (D _:k'omo_
* asteRisko
pundo
(D _pundOI_:
£ pundo
£ (D _pundoI_:
! ,ekk@-*io,
: dupunkto
- _
- (_a
$ dolaRo
$ (D _dolaROI_:
% p@-*otsento
@ po
& %kaI
$ (D _dolaRoI_:


+ 1
- 1
dictsource/es_list View File

@@ -61,7 +61,7 @@ _« kom'iJ^as||iTkJ^'ERDas
_» kom'iJ^as||dE**'EtSas

^ TiRkumfl'exo
€ eU*o
€ eU**o
% pOrTj'Ento
& ampERs'ant
@ a*R'oBa

+ 5
- 5
dictsource/hi_list View File

@@ -113,11 +113,11 @@ _97 s@t:'a:n@we:
_98 ath:'a:n@we:
_99 nIn^j'a:n@we:

_0C s'O:
_0M1 h@z'a:r
_0M2 d,Vsla:kh@_ // until we fix numbers > 100,000
_1M2 d,Vsla:kh@_ // until we fix numbers > 100,000
_0L l'a:kh
_0C s'O:
_0M1 h@z'a:r
_0M2 l'a:kh
_0M3 cr'ore
_0M4 b'ilion // ?? English
_dpt _d@s@ml'o:_ // ?? what is Hi for "." ?

// Single consonants

+ 9
- 9
dictsource/jbo_list View File

@@ -28,12 +28,12 @@ g g@
_i ibu
j Z@
k k@
l l@
m m@
n n@
_l l@
_m m@
_n n@
_o obu
p p@
r R@
_r R@
s s@
t t@
_u ubu
@@ -67,11 +67,11 @@ zo'e $u+ // pronoun
i i_: $u // sentence break (recognised by eSpeak program). Try a short pause after ".i"

// end-of-clause [_;_] before these
noi _;_nOI $u
poi _;_pOI $u
noi _;_noI $u
poi _;_poI $u
no'u _;_nohu $u
po'u _;_pohu $u
goi _;_gOI $u
goi _;_goI $u

ija _;_iZa // should this series be unstressed?
ijanai _;_iZanaI
@@ -86,7 +86,7 @@ to _::to $u // start parenthesis, pause but don't raise intonation

// terminators (include a pause after)
ku ku_:: $u
boi bOI_:: $u
boi boI_:: $u
vau vaU_:: $u
kei keI_:: $u
toi tOI_:: $u
toi toI_:: $u

+ 20
- 17
dictsource/jbo_rules View File

@@ -8,9 +8,9 @@
a a
ai aI
au aU
X) a (_ %a // single-syllable cmavo are unstressed
X) ai (_ %aI
X) au (_ %aU
X) a (_ ,a // single-syllable cmavo are unstressed
X) ai (_ ,aI
X) au (_ ,aU

.group b
b b
@@ -25,8 +25,8 @@
.group e
e e
ei eI
X) e (_ %e
X) ei (_ %eI
X) e (_ ,e
X) ei (_ ,eI

.group f
f f
@@ -42,8 +42,8 @@
.group i
i i
i (A j
X) i (_ %i
X) i (A_ %j
X) i (_ ,i
X) i (A_ ,j

.group j
j Z
@@ -52,7 +52,7 @@
k k

.group l
l l-
l l- // syllabic
A) l l
l (A l

@@ -60,15 +60,17 @@
m m

.group n
n n
n n- // syllabic
A) n n
n (A n
n (g N
n (k N

.group o
o o
oi OI
X) o (_ %o
X) oi (_ %OI
oi oI
X) o (_ ,o
X) oi (_ ,oI

.group p
p p
@@ -77,8 +79,9 @@
q kw

.group r
r r-
r r- // syllabic
A) r R
C) r (A @-*
r (A R


@@ -92,8 +95,8 @@
.group u
u u
u (A w
X) u (_ %u
X) u (A_ %w
X) u (_ ,u
X) u (A_ ,w

.group v
v v
@@ -114,9 +117,9 @@
' h
_) ' (_ '@h@

. _!
. _! // dot
_) . (@P1 _! // remove . prefix
. (_S1 _! // remove . suffix


ˈ ' // U+2c8 stress marker
ˈ ' // U+2c8 stress marker (from syllable capitalisation)

+ 11
- 10
dictsource/sw_list View File

@@ -57,15 +57,6 @@ _6C m,ias'ita
_7C m,ias'aba
_8C m,ian'ane
_9C m,iat'isa
_1L l'akim'oJa
_2L l'aki_|mb'ili
_3L l'akit'atu
_4L l'aki'n-ne
_5L l'akit'ano
_6L l'akis'ita
_7L l'akis'aba
_8L l'akin'ane
_9L l'akit'isa
_0M1 'elfu
_1M1 ,elfum'oJa
_2M1 ,elfu_|mb'ili
@@ -76,7 +67,17 @@ _6M1 ,elfus'ita
_7M1 ,elfus'aba
_8M1 ,elfun'ane
_9M1 ,elfut'isa
_0M2 _|mili'oni
_1M2 l'akim'oJa
_2M2 l'aki_|mb'ili
_3M2 l'akit'atu
_4M2 l'aki'n-ne
_5M2 l'akit'ano
_6M2 l'akis'ita
_7M2 l'akis'aba
_8M2 l'akin'ane
_9M2 l'akit'isa
_0M3 _|mili'oni // 1,000,000
_0M4 _|bili'oni
_dpt _nukta
_0and _|na


+ 4
- 0
espeak-data/voices/ca View File

@@ -0,0 +1,4 @@
name catalan-test
language ca
gender male


+ 6
- 0
espeak-data/voices/lv View File

@@ -0,0 +1,6 @@
name latvian
language lv
gender male

replace 03 o o:


+ 73
- 46
phsource/compile_report View File

@@ -1,4 +1,4 @@
58 phoneme tables
59 phoneme tables
new total
base 101 101
base2 24 120
@@ -20,6 +20,7 @@
hi 52 139
ta 17 142
hu 23 117
lv 28 123
nl 28 124
pl 18 114
sk 29 132
@@ -32,6 +33,7 @@
la 21 119
es 9 121
es_la 1 121
ca 9 122
pt 28 137
pt_pt 20 137
ro 36 144
@@ -55,9 +57,8 @@
kn 15 141
bn 59 147
ne 23 151
lv 28 123
hy 17 115
om 18 118
om 21 121

Data file Used by
b/b [b] base
@@ -292,6 +293,7 @@ h/ha [h] base
[<h>] la
[H] bn
[h] bn
[a#] om
h/he [h] base
[h] fi
[H] hi
@@ -360,6 +362,7 @@ l/l [l] base
[l] sk
[l] ru
[l] la
[l] ca
[l] pt
[l;] ro
l/_l [l] base
@@ -375,6 +378,7 @@ l/_l [l] base
[l] sk
[l] ru
[l] la
[l] ca
[l] pt
[l;] ro
[l] zh
@@ -471,6 +475,7 @@ l/tl [l] base
[l] sk
[l] ru
[l] la
[l] ca
[l] pt
[l;] ro
[l] sq
@@ -578,11 +583,11 @@ r3/r_sr [r.] bn
r3/r_trill [R2] base
[R3] base
[r] af
[R2] lv
[r] nl
[R] sk
[r*] sr
[x] pt
[R2] lv
r3/r_trill2 [R] base
[r] cy
[R] hr
@@ -593,10 +598,10 @@ r3/r_trill3.wav [R3] base
[r] af
[r] nl
r3/r_trill.wav [R2] base
[R2] lv
[R] sk
[r*] sr
[x] pt
[R2] lv
r3/r_u [(u)] base
r3/r_ulv [r"] hy
r3/r_uvl [r"] hy
@@ -714,23 +719,23 @@ ufric/s_ [s] base
[z2] fr
[z3] fr
[s2] hu
[s2] lv
[s] es
[s#] pt
[z;] ro
[s2] is
[s2] lv
ufric/s! [s] base
[s2] fi
[s] fr
[s2] hu
[s2] lv
[s] es
[s#] pt
[s2] is
[s2] lv
ufric/s_continue [s2] fi
[s2] hu
[s2] is
[s2] lv
[s2] is
ufric/sh [S] base
[S] fr
[s#] pt_pt
@@ -774,6 +779,7 @@ ufric/x_sr [h] sr
ufric/xx3 [X] base
[Q] hy
ustop/c [c] base
ustop/c_ejc [c`] om
ustop/k [k] base
[k] en
[k] fr
@@ -788,13 +794,14 @@ ustop/k_ [k] base
[k] fr
[kh] hi
[k] hu
[k] lv
[k] sk
[k] el
[kh] zh
[k] sw
[k] sq
[kh] bn
[k] lv
ustop/k_ejc [k`] om
ustop/ki [c] base
[k] base
[k] base2
@@ -803,6 +810,7 @@ ustop/ki [c] base
[k] fi
[k] fr
[k] hu
[k] lv
[k] sk
[k^] mk
[c] is
@@ -812,7 +820,7 @@ ustop/ki [c] base
[c] tr
[J] ku
[c] ku
[k] lv
ustop/ki_ejc [k`] om
ustop/kl [k] base
[k] base2
[k] en
@@ -820,12 +828,12 @@ ustop/kl [k] base
[k] fr
[kh] hi
[k] hu
[k] lv
[k] sk
[k] el
[k] zhy
[k] sw
[kh] bn
[k] lv
ustop/kr [k] base
[k] base2
[k] en
@@ -833,27 +841,27 @@ ustop/kr [k] base
[k] fr
[kh] hi
[k] hu
[k] lv
[k] sk
[k] el
[k] zhy
[k] sw
[kh] bn
[k] lv
ustop/k_unasp [k] base
[k] base2
[k] fi
[k] hi
[k] hu
[k] lv
[k] sk
[k] el
[k] zhy
[k`] ko
[k] sq
[k] bn
[k] lv
[k`] om
ustop/k_unasp_ [k] zh
ustop/null [?] base
[dh] om
ustop/p [p] base
[p] fr
[p2] fr
@@ -868,6 +876,7 @@ ustop/p_ [p] base
[ph] hy
ustop/p_asp [ph] hi
[ph] hy
ustop/p_ejc [p`] om
ustop/percus10 [(X1] base
ustop/pl [p] base
[p] base2
@@ -876,13 +885,13 @@ ustop/pl [p] base
[p] hi
[ph] hi
[p] hu
[p] lv
[p] sk
[p] la
[p;] ro
[p] sq
[p] bn
[ph] bn
[p] lv
[ph] hy
ustop/pr [p] base
[p] base2
@@ -902,6 +911,7 @@ ustop/p_unasp [p] base
[p] fr
[p] hi
[p] hu
[p] lv
[p] sk
[p] la
[p;] ro
@@ -909,18 +919,16 @@ ustop/p_unasp [p] base
[p`] ko
[p] sq
[p] bn
[p] lv
[p`] om
ustop/p_unasp_ [p] base2
[p] fi
[p] hi
[p] hu
[p] lv
[p] sk
[p] la
[p;] ro
[p] sq
[p] bn
[p] lv
ustop/q [q] base
ustop/q_u [q] base
ustop/t [t] base
@@ -935,13 +943,13 @@ ustop/t_ [t] base
[t3] fr
[t.] hi
[t.h] hi
[t] lv
[t] sk
[t] el
[th] zh
[t] sq
[t.] bn
[t.h] bn
[t] lv
ustop/t_dnt [t] base
[t[] base
[t] base2
@@ -958,8 +966,8 @@ ustop/t_dnt [t] base
[t] sw
[t`] ko
[t] bn
[t`] om
ustop/t_dnt2 [t[] vi
ustop/t_ejc [t`] om
ustop/t_hi [t] hi
[t] bn
[th] bn
@@ -980,8 +988,17 @@ ustop/ts [ts] base2
[ts;] ro
[z] zhy
[tsh] zh
[ts] hy
ustop/ts_ [ts] base2
[ts] de
[ts] eo
[ts] hu
[ts] lv
[ts] pl
[ts] ru
[ts;] ro
[ts] hy
ustop/ts2 [ts] lv
ustop/tsh [tS] base
[c] zhy
ustop/tsh_ [tS] base
@@ -992,11 +1009,11 @@ ustop/t_short [t] fr
[t2] fr
[t3] fr
[t.] hi
[t] lv
[t] sk
[t] el
[t] sq
[t.] bn
[t] lv
ustop/tsh_pzd [tS;] zh
ustop/tsh_pzd_unasp [tS;] zh
ustop/tsh_sr [tS] sr
@@ -1013,7 +1030,6 @@ ustop/ts_pzd2 [c] hi
[c] hu
[c`] ko
[c] bn
[c`] om
ustop/ts_pzd3 [tS;] hr
[dZ;] hr
ustop/t_sr [d] sr
@@ -1090,13 +1106,14 @@ vdiph2/y-y# [yY] fi
vdiph/8u [oU] en_us
[ou] zh
vdiph/aae [aI] en_us
vdiph/aai [aai] zhy
vdiph/aai [ai] lv
[aai] zhy
[aI] ne
[ai] lv
vdiph/aai_2 [AI] af
[AY] cy
vdiph/aai_3 [ai] fi
vdiph/aau [au] fi
[au] lv
vdiph/aau_2 [aU] en_wi
[aU] de
[VU] nl
@@ -1157,8 +1174,8 @@ vdiph/eei [EI] base2
[eI] hy
vdiph/eei_2 [eI] eo
[ei] fi
[eI] id
[ei] lv
[eI] id
vdiph/eei_3 [eI] sk
[eI] ku
vdiph/eeu [EU] pt
@@ -1214,18 +1231,17 @@ vdiph/ooi [OI] en
[OI] en_n
[OI] cy
[OY] cy
[OI] eo
[oI] eo
[oi] fi
[oi] lv
[OI] no
[oi] zhy
[oi] lv
vdiph/ooi_2 [OI] af
vdiph/ooi_3 [OI] en_rp
[aI] en_wm
vdiph/ooi_4 [OI] en_us
vdiph/oou [oU] cs
[OU] grc
[au] lv
vdiph/ou [ou] fi
[ou] zhy
[oU] ne
@@ -1241,10 +1257,10 @@ vdiph/@u_en [oU] en
vdiph/ui [uI] base2
[uI] eo
[ui] fi
[ui] lv
[uI] vi
[ui] zhy
[uI] ne
[ui] lv
vdiph/u-i [yI] vi
vdiph/ui_2 [uI] af
vdiph/ui_3 [uI] cy
@@ -1322,6 +1338,7 @@ voc/dh [D] base
[D] es
voc/dh_ [D] base
[D] es
voc/dh_om [d`] om
voc/j [J^] base
voc/Q [Q] base
[Q^] base
@@ -1415,11 +1432,11 @@ vowel/@- [@-] base
vowel/& [a] en_rp
[&] fi
[&:] hi
[&] lv
[&:] lv
[&] sv
[&] bn
[&:] bn
[&] lv
[&:] lv
vowel/0 [0] en
[O] hi
[O] pt
@@ -1482,13 +1499,13 @@ vowel/a# [a/] base2
[&/] pt
[a#] rw
[a/] rw
[a] om
vowel/a_2 [a] base2
[a] en_wi
[A:] en_wi
[A@] en_wi
[aa] en_wi
[a] eo
[a] ca
[A] pt
[a] pt_pt
[a] ro
@@ -1502,6 +1519,7 @@ vowel/a#_2 [a#] pl
[a/] sw
[a/] sq
[a] kn
[a] om
vowel/a_3 [a] en_sc
[a/] en_sc
[A:] en_sc
@@ -1511,7 +1529,6 @@ vowel/a_3 [a] en_sc
[a] is
[A] zh
[A] da
[a:] om
vowel/a#_3 [a2] en
[a2] en_n
[a2] en_us
@@ -1520,6 +1537,7 @@ vowel/a#_3 [a2] en
[@] de
[a] hi
[a#] ru
[a#] ca
[&] pt_pt
[&/] pt_pt
[@] vi
@@ -1528,6 +1546,7 @@ vowel/a#_3 [a2] en
[a] bn
vowel/a_4 [a/] el
[a] vi
[a:] om
vowel/a_5 [a:] ta
[a/] it
[a] la
@@ -1537,15 +1556,16 @@ vowel/a_5 [a:] ta
[a] sq
[a:] ml
[a:] kn
[A:] om
vowel/a_6 [a] fr
[a2] fr
vowel/aa [a] fi
[A] fr_ca
[a] lv
[a:] lv
[A] no
[A:] no
[aa] zhy
[a] lv
[a:] lv
[a] hy
vowel/aa# [O] en_us
[0] en_wi
@@ -1598,6 +1618,7 @@ vowel/e [e] base2
[e] hr
[e/] it
[e:] la
[e] ca
[e] pt
[e] pt_pt
[e] grc
@@ -1657,7 +1678,6 @@ vowel/ee_6 [&] sk
[E3] sv
[E] ku
vowel/e_mid [E] en_rp
[e] jbo
[E] fr
[E2] fr
[E] fr_ca
@@ -1669,6 +1689,7 @@ vowel/e_mid [E] en_rp
[E] hr
[E] it
[E] es
[E] ca
[E] pt_pt
[E] no
[E] is
@@ -1682,7 +1703,10 @@ vowel/e_mid2 [E] af
[E] de
[E2] de
[E:] de
[e] jbo
[e] fi
[e] lv
[e:] lv
[E] nl
[e] sk
[e] es
@@ -1692,8 +1716,6 @@ vowel/e_mid2 [E] af
[e] tr
[E] ko
[e] kn
[e] lv
[e:] lv
[E] om
vowel/@_fnt [@] en_wi
[@2] en_wi
@@ -1738,7 +1760,6 @@ vowel/i_3 [i] af
[i] fr
[i:] fr
[i:] sk
[i:] lv
vowel/i_4 [i] fi
[i] hu
[I] it
@@ -1749,13 +1770,14 @@ vowel/i#_5 [y] pt_pt
[i/] pt_pt
[y] ro
vowel/i_6 [i] ta
[i] lv
[i:] lv
[i] hr
[i] tr
[i] ku
[i] id
[i] ml
[i] kn
[i] lv
vowel/i#_6 [i.] zh
vowel/i_7 [i] pl
vowel/i#_7 [i[] zh
@@ -1845,6 +1867,7 @@ vowel/o [o] base2
[o] it
[o/] it
[O] la
[o] ca
[o] pt_pt
[o] grc
[o:] sv
@@ -1898,6 +1921,7 @@ vowel/oo [O:] en_sc
[O] de
[o] eo
[O] es
[O] ca
[o] el
[O] sv
[O] no
@@ -1912,13 +1936,13 @@ vowel/oo_1 [O:] en_n
[O@] en_wm
[O:] af
[o] fi
[o] lv
[o:] lv
[o] sk
[o:] sk
[o] hr
[o] vi
[O/] sq
[o] lv
[o:] lv
vowel/oo_2 [0] en_sc
[O] cy
[o] cs
@@ -1987,7 +2011,9 @@ vowel/u#_3 [U] ta
[U:] ta
vowel/u#_4 [U] en_sc
[u:] en_sc
vowel/u_5 [u] sw
vowel/u_5 [u] lv
[u:] lv
[u] sw
vowel/u_6 [U] pt_pt
[u] pt_pt
[u] ku
@@ -2026,9 +2052,8 @@ vowel/uu [U] en
vowel/uu# [U] ku
vowel/uu_2 [U] base2
[U] de
[U] ca
[U] tr
[u] lv
[u:] lv
vowel/uu_3 [u] af
[y] zh
vowel/uu_4 [U] fi
@@ -2101,7 +2126,6 @@ vowel/yy_4 [y] de
[y] is
[y] hy
vweak/@ [@#] om
vweak/a [a#] om
vwl_af/@ [@] af
vwl_af/I [I] af
vwl_af/r@ [@] af
@@ -2233,6 +2257,7 @@ vwl_zh/ung [ung] zhy
vwl_zh/uo [uo] zh
vwl_zh/y& [y&] zh
vwl_zh/yee [yE] zh
w/iw_ [w/] base
w/w [w] base
w/_w [w] base
[w] zh
@@ -2262,6 +2287,7 @@ w/wu [(u)] base
w/xw [(@)] base
[(a)] base
[(e)] base
[(i)] base
[(o)] base
[(u)] base
x/b [b] base
@@ -2281,6 +2307,7 @@ x/d [d] base
[d] is
[d] id
[d.] bn
[d`] om
x/d_ [d] base
[d[] base
[d] base2

+ 1
- 0
phsource/ph_base2 View File

@@ -161,6 +161,7 @@ phoneme ts
vowelout f1=0 f2=1700 -300 250 f3=-100 80 rms=20
lengthmod 2
wave ustop/ts
before _ ustop/ts_
endphoneme



+ 67
- 0
phsource/ph_catalan View File

@@ -0,0 +1,67 @@

//====================================================
// Catalan - based on Spanish and Base2
//====================================================

phoneme a
vowel starttype (a) endtype (a)
length 180
formants vowel/a_2
reduceto a# 2
endphoneme

phoneme a#
vowel starttype (@) endtype (@)
length 170
formants vowel/a#_3
unstressed
endphoneme

phoneme e
vowel starttype (e) endtype (e)
length 170
formants vowel/e
reduceto @ 2
endphoneme

phoneme E
vowel starttype (e) endtype (e)
length 180
formants vowel/e_mid
reduceto @ 2
endphoneme

phoneme O
vowel starttype (o) endtype (o)
length 170
formants vowel/oo
reduceto U 2
endphoneme

phoneme o
vowel starttype (o) endtype (o)
length 170
formants vowel/o
reduceto U 2
endphoneme

phoneme U
vowel starttype (u) endtype (u)
length 160
formants vowel/uu_2
unstressed
endphoneme



phoneme l
liquid
length 100
lengthmod 7
beforenotvowel l/2 // use 'dark' [l] after a vowel
formants l/l
after _ l/_l t l/tl
endphoneme




+ 8
- 7
phsource/ph_esperanto View File

@@ -42,42 +42,42 @@ endphoneme

phoneme aU
vowel starttype (a) endtype (u)
length 230
length 250
formants vdiph/au_4
endphoneme


phoneme eU
vowel starttype (e) endtype (u)
length 230
length 250
formants vdiph/eu
endphoneme


phoneme aI
vowel starttype (a) endtype (i)
length 240
length 250
formants vdiph/ai
endphoneme


phoneme eI
vowel starttype (e) endtype (i)
length 230
length 250
formants vdiph/eei_2
endphoneme


phoneme OI
phoneme oI
vowel starttype (o) endtype (i)
length 240
length 250
formants vdiph/ooi
endphoneme


phoneme uI
vowel starttype (u) endtype (i)
length 230
length 250
formants vdiph/ui
endphoneme

@@ -93,6 +93,7 @@ phoneme ts
vowelout f1=0 f2=1700 -300 250 f3=-100 80 rms=20
lengthmod 2
wave ustop/ts
before _ ustop/ts_
endphoneme



+ 1
- 0
phsource/ph_german View File

@@ -218,6 +218,7 @@ phoneme ts
vowelout f1=0 f2=1700 -300 250 f3=-100 80 rms=20
lengthmod 2
wave ustop/ts
before _ ustop/ts_
endphoneme



+ 1
- 0
phsource/ph_hungarian View File

@@ -102,6 +102,7 @@ phoneme ts
vowelout f1=0 f2=1700 -300 250 f3=-100 80 rms=20
lengthmod 2
wave ustop/ts
before _ ustop/ts_
endphoneme



+ 217
- 0
phsource/ph_latvian View File

@@ -0,0 +1,217 @@

phoneme @
vowel starttype (@) endtype (@)
length 120
formants vowel/8_7
endphoneme
phoneme i
vowel starttype (i) endtype (i)
length 110
formants vowel/i_6
linkout ;
endphoneme

phoneme i:
vowel starttype (i) endtype (i)
length 300
formants vowel/i_6
linkout ;
endphoneme


phoneme e
vowel starttype (e) endtype (e)
length 120
formants vowel/e_mid2
endphoneme

phoneme e:
vowel starttype (e) endtype (e)
length 330
formants vowel/e_mid2
endphoneme


phoneme &
vowel starttype (a) endtype (a)
length 120
formants vowel/&
endphoneme

phoneme &:
vowel starttype (a) endtype (a)
length 330
formants vowel/&
endphoneme


phoneme a
vowel starttype (a) endtype (a)
length 120
formants vowel/aa
endphoneme

phoneme a:
vowel starttype (a) endtype (a)
length 350
formants vowel/aa
endphoneme


phoneme o
vowel starttype (o) endtype (o)
length 120
formants vowel/oo_1
endphoneme

phoneme o:
vowel starttype (o) endtype (o)
length 330
formants vowel/oo_1
endphoneme


phoneme u
vowel starttype (u) endtype (u)
length 110
formants vowel/u_5
endphoneme

phoneme u:
vowel starttype (u) endtype (u)
length 330
formants vowel/u_5
endphoneme


phoneme ai
vowel starttype (a) endtype (i)
length 310
long
formants vdiph/aai
endphoneme

phoneme ei
vowel starttype (e) endtype (i)
length 300
long
formants vdiph/eei_2
endphoneme

phoneme au
vowel starttype (a) endtype (u)
length 300
long
formants vdiph/aau
endphoneme

phoneme uo
vowel starttype (u) endtype (a)
length 300
long
// formants vdiph2/uo
formants vdiph2/uaa
endphoneme

phoneme oi
vowel starttype (o) endtype (i)
length 290
long
formants vdiph/ooi
endphoneme

phoneme ui
vowel starttype (u) endtype (i)
length 290
long
formants vdiph/ui
endphoneme

phoneme ie
vowel starttype (i) endtype (e)
length 290
long
formants vdiph2/ie_2
endphoneme

phoneme iu
vowel starttype (i) endtype (u)
length 290
long
formants vdiph2/iu_2
endphoneme




phoneme R2 // this is [R] from Slovak/Czech
liquid
vowelin f1=0 f2=1700 -300 300 f3=-300 80
vowelout f1=2 f2=1700 -300 300 f3=-300 80 brk
formants r3/r_trill+r3/r_trill.wav%50
trill
lengthmod 6
endphoneme


phoneme ts
vls alv afr sibilant lengthenstop
vowelin f1=0 f2=1700 -300 300 f3=-100 80
vowelout f1=0 f2=1700 -300 250 f3=-100 80 rms=20
lengthmod 2
wave ustop/ts2%90
before _ ustop/ts_
endphoneme


phoneme t // try disable this and try using English [t]
vls alv stop lengthenstop
vowelin f1=0 f2=1700 -300 300 f3=-100 80
vowelout f1=0 f2=1700 -300 250 f3=-100 80 rms=20
lengthmod 2
wave ustop/t_short%115
before _ ustop/t_
switchvoicing d
endphoneme

phoneme p
vls blb stop
vowelin f1=0 f2=1000 -50 -100 f3=-200 80
vowelout f1=0 f2=1000 -500 -350 f3=-300 80 rms=30
lengthmod 2
wave ustop/p_unasp%120
before _ ustop/p_unasp_%80
before l ustop/pl
switchvoicing b
endphoneme


phoneme k
vls vel stop
vowelin f1=0 f2=2300 200 400 f3=-100 80
vowelout f1=0 f2=2300 300 400 f3=-100 80 rms=20
lengthmod 2
wave ustop/k_unasp%60 // weaker
before _ ustop/k_
before r ustop/kr
before l ustop/kl
before (i) ustop/ki
before l^ ustop/ki
switchvoicing g
endphoneme


phoneme s2 // second part of long [s:]
vls alv frc sibilant
vowelin f1=0 f2=1700 -300 300 f3=-100 80
vowelout f1=0 f2=1700 -300 250 f3=-100 80 rms=20
lengthmod 3
wave ufric/s_continue
before _ ufric/s_
before p ufric/s!
before t ufric/s!
before k ufric/s!
switchvoicing z
endphoneme


+ 1
- 1
phsource/ph_lojban View File

@@ -17,7 +17,7 @@ endphoneme
phoneme e
vowel starttype (e) endtype (e)
length 170
formants vowel/e_mid
formants vowel/e_mid2
endphoneme



+ 1
- 0
phsource/ph_polish View File

@@ -135,6 +135,7 @@ phoneme ts
vowelout f1=0 f2=1700 -300 250 f3=-100 80 rms=20
lengthmod 2
wave ustop/ts
before _ ustop/ts_
endphoneme



+ 1
- 0
phsource/ph_romanian View File

@@ -219,6 +219,7 @@ phoneme ts;
vowelout f1=0 f2=1700 -300 250 f3=-100 80 rms=20 colr=1
lengthmod 2
wave ustop/ts
before _ ustop/ts_
endphoneme



+ 1
- 0
phsource/ph_russian View File

@@ -40,6 +40,7 @@ phoneme ts
vowelout f1=0 f2=1700 -300 250 f3=-100 80 rms=20
lengthmod 2
wave ustop/ts
before _ ustop/ts_
endphoneme



+ 1
- 1
phsource/ph_tamil View File

@@ -83,7 +83,7 @@ endphoneme

phoneme u:
vowel starttype (u) endtype (u)
length 250
length 260
formants vowel/u
endphoneme


+ 10
- 4
phsource/phonemes View File

@@ -115,7 +115,7 @@ phoneme _| // Very short pause
formants NULL
starttype _ endtype _
length 1
lengthmod 1
lengthmod 5
endphoneme

phoneme _: // Standard pause
@@ -301,6 +301,7 @@ phoneme (i) virtual
before N DFT-40+nn/inn
before ; DFT-40+j2/xj2
before l DFT-40+l/xl
before w DFT-30+w/xw
before j DFT-30+j/xj
before r- DFT-60+r/xr
before r/ DFT+r3/r_i
@@ -498,6 +499,7 @@ phoneme w/ // used for [w] when not before a vowel
vowelout len=50
length 100
formants w/w_
after (i) w/iw_
lengthmod 7
beforevowel w
endphoneme
@@ -1224,6 +1226,9 @@ include ph_tamil
phonemetable hu base
include ph_hungarian

phonemetable lv base
include ph_latvian

phonemetable nl base
include ph_dutch

@@ -1261,6 +1266,9 @@ include ph_spanish
phonemetable es_la es
include ph_spanish_la

phonemetable ca es
include ph_catalan

phonemetable pt base2
include ph_pt_brazil

@@ -1333,11 +1341,9 @@ include ph_bengali
phonemetable ne hi
include ph_nepali

phonemetable lv base
include ph_latvian

phonemetable hy base
include ph_armenian_west

phonemetable om base
include ph_oromo


BIN
phsource/ustop/ts2.wav View File


BIN
phsource/ustop/ts_.wav View File


BIN
phsource/vdiph/aai View File


BIN
phsource/vdiph/aau View File


BIN
phsource/vdiph/eei_2 View File


BIN
phsource/vdiph/vowelchart.png View File


BIN
phsource/vowel/aa View File


BIN
phsource/vowel/i_3 View File


BIN
phsource/vowel/i_6 View File


BIN
phsource/w/w_ View File


+ 2
- 1
src/compiledata.cpp View File

@@ -264,6 +264,7 @@ static keywtab_t keywords[] = {
{"long", 0x2000000+phLONG},
{"brkafter", 0x2000000+phBRKAFTER},
{"nonsyllabic",0x2000000+phNONSYLLABIC},
{"lengthenstop",0x2000000+phLENGTHENSTOP},

// voiced / unvoiced
{"vcd", 0x2000000+phVOICED},
@@ -602,7 +603,7 @@ int Compile::LoadSpect(const char *path, int control)
float total;
float pkheight;
int marker1_set=0;
int frame_vowelbreak=NULL;
int frame_vowelbreak=0;
SpectFrame *fr;
wxString path_sep = _T("/");


+ 25
- 9
src/compiledict.cpp View File

@@ -50,6 +50,7 @@ static int debug_flag = 0;

int hash_counts[N_HASH_DICT];
char *hash_chains[N_HASH_DICT];
char letterGroupsDefined[N_LETTER_GROUPS];

MNEM_TAB mnem_flags[] = {
// these in the first group put a value in bits0-3 of dictionary_flags
@@ -199,6 +200,10 @@ static char nullstring[] = {0};
text_not_phonemes = 0;
phonetic = word = nullstring;

if(memcmp(linebuf,"_-",2)==0)
{
step=1; // TEST
}
p = linebuf;
// while(isspace2(*p)) p++;

@@ -313,7 +318,7 @@ static char nullstring[] = {0};
break;
case 1:
if(c == '-')
if((c == '-') && (word[0] != '_'))
{
flag_codes[n_flag_codes++] = BITNUM_FLAG_HYPHENATED;
c = ' ';
@@ -798,10 +803,16 @@ void copy_rule_string(char *string, int &state)
c = *p++ - '0';
value = *p++ - '0';
c = c * 10 + value;
if((value < 0) || (value > 9) || (c <= 0) || (c >= N_LETTER_GROUPS))
if((value < 0) || (value > 9))
{
c = 0;
fprintf(f_log,"%5d: Expected 2 digits after 'L'",linenum);
fprintf(f_log,"%5d: Expected 2 digits after 'L'\n",linenum);
error_count++;
}
else
if((c <= 0) || (c >= N_LETTER_GROUPS) || (letterGroupsDefined[(int)c] == 0))
{
fprintf(f_log,"%5d: Letter group L%.2d not defined\n",linenum,c);
error_count++;
}
c += 'A';
@@ -1249,18 +1260,25 @@ static int compile_lettergroup(char *input, FILE *f_out)
p = input;
if(!isdigit(p[0]) || !isdigit(p[1]))
{
fprintf(f_log,"%5d: Expected 2 digits after '.L'\n",linenum);
error_count++;
return(1);
}

group = atoi(&p[1]);
group = atoi(&p[0]);
if(group >= N_LETTER_GROUPS)
{
fprintf(f_log,"%5d: lettergroup out of range (01-%.2d)\n",linenum,N_LETTER_GROUPS);
error_count++;
return(1);
}

while(!isspace2(*p)) p++;

fputc(RULE_GROUP_START,f_out);
fputc(RULE_LETTERGP2,f_out);
fputc(group + 'A', f_out);
letterGroupsDefined[group] = 1;

for(;;)
{
@@ -1346,11 +1364,7 @@ static int compile_dictrules(FILE *f_in, FILE *f_out, char *fname_temp)

if(memcmp(buf,".L",2)==0)
{
if(compile_lettergroup(&buf[2], f_out) != 0)
{
fprintf(f_log,"%5d: Bad lettergroup\n",linenum);
error_count++;
}
compile_lettergroup(&buf[2], f_out);
continue;
}

@@ -1509,6 +1523,8 @@ int CompileDictionary(const char *dsource, const char *dict_name, FILE *log, cha
char path[sizeof(path_home)+40]; // path_dsource+20

error_count = 0;
memset(letterGroupsDefined,0,sizeof(letterGroupsDefined));

debug_flag = flags & 1;

if(dsource == NULL)

+ 13
- 1
src/dictionary.cpp View File

@@ -686,6 +686,8 @@ int Translator::IsLetterGroup(char *word, int group, int pre)
char *w;

p = letterGroups[group];
if(p == NULL)
return(0);

while(*p != RULE_GROUP_END)
{
@@ -1012,7 +1014,9 @@ void Translator::SetWordStress(char *output, unsigned int dictionary_flags, int
int unstressed_word = 0;
char *max_output;
int final_ph;
int final_ph2;
int mnem;
int mnem2;
int post_tonic;
int opt_length;
int done;
@@ -1043,6 +1047,7 @@ void Translator::SetWordStress(char *output, unsigned int dictionary_flags, int
}
if(ix == 0) return;
final_ph = phonetic[ix-1];
final_ph2 = phonetic[ix-2];

max_output = output + (N_WORD_PHONEMES-3); /* check for overrun */

@@ -1131,7 +1136,14 @@ void Translator::SetWordStress(char *output, unsigned int dictionary_flags, int
else
{
mnem = phoneme_tab[final_ph]->mnemonic;
if((mnem != 'n') && (mnem != 's'))
mnem2 = phoneme_tab[final_ph2]->mnemonic;

if((mnem == 's') && (mnem2 == 'n'))
{
// -ns stress remains on penultimate syllable
}
else
if(((mnem != 'n') && (mnem != 's')) || (phoneme_tab[final_ph2]->type != phVOWEL))
{
stressed_syllable = vowel_count - 1;
}

+ 12
- 20
src/numbers.cpp View File

@@ -756,6 +756,7 @@ int Translator::TranslateRoman(char *word, char *ph_out)
int subtract;
int repeat = 0;
unsigned int flags;
char ph_roman[30];
char number_chars[N_WORD_BYTES];

static const char *roman_numbers = "ixcmvld";
@@ -806,11 +807,20 @@ int Translator::TranslateRoman(char *word, char *ph_out)
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)];
Lookup("_roman",ph_roman); // precede by "roman" if _rom is defined in *_list
p = &ph_out[0];

if((langopts.numbers & NUM_ROMAN_AFTER) == 0)
{
strcpy(ph_out,ph_roman);
p = &ph_out[strlen(ph_out)];
}

sprintf(number_chars," %d ",acc);
TranslateNumber(&number_chars[1],p,&flags,0);

if(langopts.numbers & NUM_ROMAN_AFTER)
strcat(ph_out,ph_roman);
return(1);
} // end of TranslateRoman

@@ -1236,24 +1246,6 @@ int Translator::TranslateNumber_1(char *word, char *ph_out, unsigned int *flags,
{
if((thousandplex > 0) && (value < 1000))
{
if(langopts.numbers2 & 0x100)
{
if((thousandplex == 1) && (value >= 100))
{
// special word for 100,000's
char ph_buf3[20];
sprintf(string,"_%dL",value / 100);
if(Lookup(string,ph_buf2) == 0)
{
LookupNum2(value/100,0,ph_buf2);
Lookup("_0L",ph_buf3);
strcat(ph_buf2,ph_buf3);
}
value %= 100;
if(value == 0)
suppress_null = 1;
}
}
if((suppress_null == 0) && (LookupThousands(value,thousandplex,ph_append)))
{
// found an exact match for N thousand

+ 3
- 2
src/phoneme.h View File

@@ -56,8 +56,9 @@
#define phBEFORENOTVOWEL2 0x1000
#define phSWITCHVOICING 0x0800

#define phNONSYLLABIC 0x100000 // don't count this vowel as a syllable when finding the stress position
#define phLONG 0x200000
#define phNONSYLLABIC 0x100000 // don't count this vowel as a syllable when finding the stress position
#define phLONG 0x200000
#define phLENGTHENSTOP 0x400000 // make the pre-pause slightly longer

// fixed phoneme code numbers, these can be used from the program code
#define phonCONTROL 1

+ 5
- 1
src/phonemelist.cpp View File

@@ -557,7 +557,11 @@ void Translator::MakePhonemeList(int post_pause, int start_sentence)
{
if((x = (langopts.word_gap & 0x7)) != 0)
{
insert_ph = pause_phonemes[x];
if((x > 1) || ((insert_ph != phonPAUSE_SHORT) && (insert_ph != phonPAUSE_NOLINK)))
{
// don't reduce the pause
insert_ph = pause_phonemes[x];
}
}
if(option_wordgap > 0)
{

+ 5
- 2
src/setlengths.cpp View File

@@ -320,6 +320,9 @@ void Translator::CalcLengths()
if((langopts.word_gap & 0x10) && (p->newword))
p->prepause = 60;

if(p->ph->phflags & phLENGTHENSTOP)
p->prepause += 30;

if(p->synthflags & SFLAG_LENGTHEN)
p->prepause += langopts.long_stop;
break;
@@ -487,9 +490,9 @@ void Translator::CalcLengths()
}

// calc length modifier
if(next->ph->code == phonPAUSE_VSHORT)
if((next->ph->code == phonPAUSE_VSHORT) && (next2->type == phPAUSE))
{
// ignore very short pause
// if PAUSE_VSHORT is followed by a pause, then use that
next = next2;
next2 = next3;
next3 = &phoneme_list[ix+4];

+ 1
- 1
src/synthdata.cpp View File

@@ -35,7 +35,7 @@
#include "translate.h"
#include "wave.h"

const char *version_string = "1.39.22 10.Nov.08";
const char *version_string = "1.39.26 15.Nov.08";
const int version_phdata = 0x013900;

int option_device_number = -1;

+ 35
- 8
src/tr_languages.cpp View File

@@ -275,7 +275,7 @@ Translator *SelectTranslator(const char *name)

case L('e','o'):
{
static const short stress_lengths_eo[8] = {145, 145, 200, 170, 0, 0, 320, 340};
static const short stress_lengths_eo[8] = {145, 145, 230, 170, 0, 0, 360, 370};
static const unsigned char stress_amps_eo[] = {16,14, 20,20, 20,24, 24,22 };
static const wchar_t eo_char_apostrophe[2] = {'l',0};
@@ -286,6 +286,7 @@ Translator *SelectTranslator(const char *name)
tr->char_plus_apostrophe = eo_char_apostrophe;

tr->langopts.word_gap = 1;
tr->langopts.vowel_pause = 2;
tr->langopts.stress_rule = 2;
tr->langopts.stress_flags = 0x6 | 0x10;
tr->langopts.unstressed_wd1 = 3;
@@ -296,9 +297,12 @@ Translator *SelectTranslator(const char *name)
break;

case L('e','s'): // Spanish
case L('c','a'): // Catalan
{
static const short stress_lengths_es[8] = {170, 200, 180, 180, 0, 0, 220, 250};
static const short stress_lengths_es[8] = {180, 210, 190, 190, 0, 0, 230, 260};
// static const short stress_lengths_es[8] = {170, 200, 180, 180, 0, 0, 220, 250};
static const unsigned char stress_amps_es[8] = {16,12, 18,18, 20,20, 20,20 }; // 'diminished' is used to mark a quieter, final unstressed syllable
static const wchar_t ca_punct_within_word[] = {'\'',0xb7,0}; // ca: allow middle-dot within word

tr = new Translator();
SetupTranslator(tr,stress_lengths_es,stress_amps_es);
@@ -313,7 +317,13 @@ Translator *SelectTranslator(const char *name)
tr->langopts.unstressed_wd2 = 2;
tr->langopts.param[LOPT_SONORANT_MIN] = 120; // limit the shortening of sonorants before short vowels

tr->langopts.numbers = 0x529 + NUM_ROMAN;
tr->langopts.numbers = 0x529 + NUM_ROMAN + NUM_ROMAN_AFTER;

if(name2 == L('c','a'))
{
tr->punct_within_word = ca_punct_within_word;
tr->langopts.stress_flags = 0x200 | 0x6 | 0x30; // stress last syllable unless word ends with a vowel
}
}
break;

@@ -369,7 +379,7 @@ Translator *SelectTranslator(const char *name)
tr->langopts.stress_rule = 6; // stress on last heaviest syllable, excluding final syllable
tr->langopts.stress_flags = 0x10004; // use 'diminished' for unstressed final syllable
tr->langopts.numbers = 0x011;
tr->langopts.numbers2 = 0x100;
tr->langopts.numbers2 = NUM2_100000;
tr->letter_bits_offset = OFFSET_DEVANAGARI;
SetIndicLetters(tr);
}
@@ -515,12 +525,13 @@ SetLengthMods(tr,3); // all equal

case L_jbo: // Lojban
{
static const short stress_lengths_jbo[8] = {185,170, 200,200, 0,0, 290,300};
static const short stress_lengths_jbo[8] = {145,145, 170,160, 0,0, 330,350};
static const wchar_t jbo_punct_within_word[] = {'.',',','\'',0x2c8,0}; // allow period and comma within a word, also stress marker (from LOPT_SYLLABLE_CAPS)
tr = new Translator();
SetupTranslator(tr,stress_lengths_jbo,NULL);
tr->langopts.stress_rule = 2;
tr->langopts.vowel_pause = 0x20c; // pause before a word which starts with a vowel, or after a word which ends in a consonant
// tr->langopts.word_gap = 1;
tr->punct_within_word = jbo_punct_within_word;
tr->langopts.param[LOPT_SYLLABLE_CAPS] = 1; // capitals indicate stressed syllables
SetLetterVowel(tr,'y');
@@ -577,8 +588,10 @@ SetLengthMods(tr,3); // all equal

case L('l','v'): // latvian
{
static const unsigned char stress_amps_lv[8] = {17,14, 20,20, 20,22, 22,22 };
static const short stress_lengths_lv[8] = {180,140, 210,210, 0,0, 210,210};
tr = new Translator();
SetupTranslator(tr,stress_lengths_sk,stress_amps_sk);
SetupTranslator(tr,stress_lengths_lv,stress_amps_lv);

tr->langopts.stress_rule = 0;
tr->langopts.spelling_stress = 1;
@@ -636,6 +649,17 @@ SetLengthMods(tr,3); // all equal
}
break;

case L('o','m'):
{
static const unsigned char stress_amps_om[] = {16,16, 20,20, 20,24, 24,22 };
static const short stress_lengths_om[8] = {200,200, 200,200, 0,0, 200,200};
tr = new Translator();
SetupTranslator(tr,stress_lengths_om,stress_amps_om);

tr->langopts.stress_rule = 3;
}
break;

case L('p','l'): // Polish
{
static const short stress_lengths_pl[8] = {160, 190, 175, 175, 0, 0, 200, 210};
@@ -780,7 +804,7 @@ SetLengthMods(tr,3); // all equal
tr->langopts.stress_flags = 0x6 | 0x10;

tr->langopts.numbers = 0x4e1;
tr->langopts.numbers2 = 0x100;
tr->langopts.numbers2 = NUM2_100000a;
}
break;

@@ -807,13 +831,15 @@ SetLengthMods(tr,3); // all equal
if(name2 == L('k','n'))
{
tr->letter_bits_offset = OFFSET_KANNADA;
tr->langopts.numbers = 0x1;
tr->langopts.numbers2 = NUM2_100000;
}
tr->langopts.param[LOPT_WORD_MERGE] = 1; // don't break vowels betwen words
SetIndicLetters(tr); // call this after setting OFFSET_
}
break;

#ifdef deleted
case L('t','h'): // Thai
{
static const short stress_lengths_th[8] = {230,150, 230,230, 230,0, 230,250};
@@ -830,6 +856,7 @@ SetLengthMods(tr,3); // all equal
tr->langopts.word_gap = 0x21; // length of a final vowel is less dependent on the next consonant, don't merge consonant with next word
}
break;
#endif

case L('t','r'): // Turkish
{

+ 62
- 6
src/translate.cpp View File

@@ -2597,10 +2597,14 @@ if((c == '/') && (langopts.testing & 2) && IsDigit09(next_in) && IsAlpha(prev_ou

for(ix=0; ix<word_count; ix++)
{
int j;
int nx;
int c_temp;
char *pn;
char *pw;
static unsigned int break_numbers1 = 0x49249248;
static unsigned int break_numbers2 = 0x492492a8; // for languages which have numbers for 100,000 and 100,00,000
static unsigned int break_numbers3 = 0x49249268; // for languages which have numbers for 100,000 and 1,000,000
unsigned int break_numbers;
char number_buf[80];

// start speaking at a specified word position in the text?
@@ -2617,24 +2621,76 @@ if((c == '/') && (langopts.testing & 2) && IsDigit09(next_in) && IsAlpha(prev_ou

// digits should have been converted to Latin alphabet ('0' to '9')
word = pw = &sbuf[words[ix].start];

if(iswdigit(word[0]) && (langopts.numbers2 & NUM2_100000))
{
// Languages with 100000 numbers. Remove thousands separators so that we can insert them again later
pn = number_buf;
while(pn < &number_buf[sizeof(number_buf)-3])
{
if(iswdigit(*pw))
{
*pn++ = *pw++;
}
else
if((*pw == langopts.thousands_sep) && (pw[1] == ' ') && iswdigit(pw[2]))
{
pw += 2;
ix++; // skip "word"
}
else
{
nx = pw - word;
memset(word,' ',nx);
nx = pn - number_buf;
memcpy(word,number_buf,nx);
break;
}
}
pw = word;
}

for(n_digits=0; iswdigit(word[n_digits]); n_digits++); // count consecutive digits

if((n_digits > 4) && (word[0] != '0'))
{
// word is entirely digits, insert commas and break into 3 digit "words"
number_buf[0] = ' ';
pn = &number_buf[1];
j = n_digits;
nx = n_digits;

if(langopts.numbers2 & NUM2_100000a)
break_numbers = break_numbers3;
else
if(langopts.numbers2 & NUM2_100000)
break_numbers = break_numbers2;
else
break_numbers = break_numbers1;

while(pn < &number_buf[sizeof(number_buf)-3])
{
if(!isdigit(c = *pw++) && (c != langopts.decimal_sep))
break;

*pn++ = c;
if((--j > 0) && (j % 3)==0)
if((--nx > 0) && (break_numbers & (1 << nx)))
{
if(langopts.thousands_sep != ' ')
{
*pn++ = langopts.thousands_sep;
}
*pn++ = ' ';
if(break_numbers & (1 << (nx-1)))
{
// the next group only has 1 digits (i.e. NUM2_10000), make it three
*pn++ = '0';
*pn++ = '0';
}
if(break_numbers & (1 << (nx-2)))
{
// the next group only has 2 digits (i.e. NUM2_10000), make it three
*pn++ = '0';
}
}
}
pn[0] = ' ';
@@ -2659,10 +2715,10 @@ if((c == '/') && (langopts.testing & 2) && IsDigit09(next_in) && IsAlpha(prev_ou
for(pw = word; *pw != ' ';)
{
memset(number_buf,' ',9);
j = utf8_in(&c_temp, pw, 0);
memcpy(&number_buf[2],pw,j);
nx = utf8_in(&c_temp, pw, 0);
memcpy(&number_buf[2],pw,nx);
TranslateWord2(&number_buf[2], &words[ix], 0, 0 );
pw += j;
pw += nx;
}
}


+ 10
- 5
src/translate.h View File

@@ -29,7 +29,7 @@
#define N_RULE_GROUP2 120 // max num of two-letter rule chains
#define N_HASH_DICT 1024
#define N_CHARSETS 20
#define N_LETTER_GROUPS 20
#define N_LETTER_GROUPS 25


/* dictionary flags, word 1 */
@@ -322,9 +322,10 @@ typedef struct {
unsigned char *length_mods;
unsigned char *length_mods0;

#define NUM_ROMAN 0x20000
#define NUM_ROMAN_UC 0x40000
#define NUM_NOPAUSE 0x80000
#define NUM_ROMAN 0x20000
#define NUM_ROMAN_UC 0x40000
#define NUM_NOPAUSE 0x80000
#define NUM_ROMAN_AFTER 0x200000

// bits0-1=which numbers routine to use.
// bit2= thousands separator must be space
@@ -344,13 +345,17 @@ typedef struct {
// bit18=Roman numbers only if upper case
// bit19=don't add pause after a number
// bit20='and' before hundreds
// bit21= say "roman" after the number, not before
int numbers;

#define NUM2_100000 0x100 // numbers for 100,000 and 10,000,000
#define NUM2_100000a 0x500 // numbers for 100,000 and 1,000,000
// 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
// bit8=(LANG=hi) special word for 100,000 and 10,000,000
// bit9=(LANG=rw) say "thousand" and "million" before its number, not after
// bit10=(LANG=sw) special word for 100,000 and 1,000,000
int numbers2;

int max_roman;

Write Preview
Loading…
Cancel
Save