mjohnson/espeak-ng
1
0
Fork 0
mirror of https://github.com/rhasspy/espeak-ng.git synced 2026-11-01 21:58:52 +00:00
Code Issues Packages Projects Releases Wiki Activity

[1.41.15]

Browse Source 
Fixes to: announce punctuation ( --punct option)


git-svn-id: https://espeak.svn.sourceforge.net/svnroot/espeak/trunk@227 d46cf337-b52f-0410-862d-fd96e6ae7743
This commit is contained in:
jonsd
2009-10-12 16:38:17 +00:00
parent 749c336768
commit 1ef2aa6091
27 changed files with 365 additions and 130 deletions
Download Patch File Download Diff File Expand all files Collapse all files
dictsource/dict_phonemes
+9 -2
View File
@@ -595,11 +595,18 @@ t tS v w x z
Dictionary pa_dict
@ a a~ e E e~ E~ i
I i: i~ I~ o O o: O~
o~ r- U u u~ U~ V V~
I i~ I~ o O O~ o~ r-
U u u~ U~ V V~
: ; b bh c ch d d.
dh dh. f g gh H j J
Jh k kh l l. m n N
n. n^ p ph Q R s S
t t. th th. v x z
Dictionary prs_dict
@ A i o
b d h l t X
dictsource/en_list
+4 -3
View File
@@ -262,8 +262,8 @@ U+32f n,0nsI2l'abI2k
// numeric
_0 zero $text // TEST
//_0 z'i@roU
//_0 zero $text // TEST
_0 z'i@roU
_1 w'0n
?6 _1 w'Vn
_2 t'u:
@@ -1582,12 +1582,13 @@ offers 0f3z
ogre oUg3
okay oUk'eI
oled oUlEd // abbrev
omeg oUm'eIg@
omega oUm'eIg@
omelet 0ml@t
omelette 0ml@t
omnibus $1
omniscience 0mn'IsI@ns
opaque oUp'eIk
oped 0p'Ed
opine oUp'aIn
opossum @p0s@m
opponent @poUn@nt
dictsource/fr_list
+3 -1
View File
@@ -24,7 +24,7 @@
// 2006-11-18 Gilles Casse <gcasse@oralux.org>
//
// Updated 2009-09-28 Michel Such <michel.such@free.fr>
// Updated 2009-10-09 Michel Such <michel.such@free.fr>
//
// * Numbers, a few abbreviations and exceptions.
//
@@ -864,11 +864,13 @@ vincent vE~sA~
afrikaans afrikans
bahamas baamas
bangladesh bA~gladES
bayeux bajY
bélarus belarys
belize beliz
bissau bisa'o
(burkina faso) byrkina||faso
burundi burundi
calvados calvadoss $text
croatie kroasi
esperanto E2sperA~to
groenland groEnlA~d
dictsource/fr_rules
+3 -1
View File
@@ -23,7 +23,7 @@
// 2006-11-18 Gilles Casse <gcasse@oralux.org>
//
// Updated: 2009-09-25 Michel Such <michel.such@free.fr>
// Updated: 2009-10-09 Michel Such <michel.such@free.fr>
//
// Letter classes:
@@ -300,6 +300,7 @@ _) en (A A~n // enivrer
ol) en (t_ A~ // somnolent, violent
Vol) en (t_ // somnolent, violent
Xol) ent (_ t2 // volent
eul) ent (_ t2 // veulent
ul) en (t_ A~ // turbulent
Vul) en (t_
imCoC) en (t_ A~ // impotent
@@ -878,6 +879,7 @@ C) i (es_ i
_) n (_° nymero
_) neuro (@@P5 nYro
_) nvda (P4 Envedea
.group o
dictsource/hu_list
+4 -1
View File
@@ -49,7 +49,8 @@ _0Z1 tizEd // tenths
_0Z2 sa:zAd // hundredths
_0Z3 EzR2Ed
_0Z4 ti:zEzR2Ed
_0Z5 sa:zEzR2Ed
_0Z6 miliomod
// ordinal numbers
@@ -90,6 +91,8 @@ szeptember $alt
október $alt
november $alt
december $alt
//The following 12 row contains shorter hungarian month names
jan $alt
feb $alt
dictsource/hu_rules
+40 -2
View File
@@ -16,12 +16,15 @@ _) alattvaló _!'AlAtvAlo:
.group á
á a:
@) ában (_S4 a:bAn
@) ának (_S4 a:nAk
.group b
b b
bb b:
@) ban (_S3 bAn
@) ben (_S3 bEn
biz (A b'iz, // bizalmas etc.
biedermeier bi:dER2ma:jER2
@@ -75,14 +78,20 @@ pá) c (só ts
dd d:
dts tS:
dt t:
apró) d (sz d
a) dsz ts:
beszé) d (szin d
cselé) d d
enge) dsz ts:
engedelmeske) dsz ts:
fogó) dz (kod ts
hazu) dsz ts:
ha) d (járat d
hor) d (szék d
mara) dsz ts:
tu) dsz ts:
kére) dz ts
mentegető) dz z
// dd d:
dz dz
A) dz (A d:z
@@ -105,11 +114,14 @@ min) dny n^
.group e
e E
D_-_) es (_ %ES
_) egyezség _!'EJ:ESSe:g
.group é
é e:
éi (é e::i:
@) ében (_S4 e:bEn
@) ének (_S4 e:nEk
.group ë
ë Y
@@ -131,11 +143,14 @@ gg g:
C) gyj J
A) gyj (A J:
ha) gyj J:
_e) gy (es J:
e) gy (jelentésű J
_e) gy (es J:
_e) gy (et_ J:
_e) gy (etlen J:
_e) gy (ik J:
_e) gy (üt J:
e) gyj (elentésű Jj
gyön) gy (sor J
_minde) gy (ik J:
na) gysz (a Js
@@ -155,17 +170,25 @@ le) g (gy g
.group i
i i
@) i (_S1 i
_) itthon ithon
.group í
í i:
íts i:tS:
fesz) ítsd i:tSd
bizony) ítsd i:tSd
.group j
j j
@) jában (_S5 ja:bAn
@) jának (_S5 ja:nAk
_) javíts jAvi:tS:
_) javítsd jAvi:tSd
.group k
k k
kk k:
@@ -271,6 +294,7 @@ szts stS:
föld) s (ánc S
s (színű S
hel) s (inki z
ki) s (ebb SS
_má) s (szó S
_munká) s S
@@ -313,12 +337,18 @@ ttn t:n
tty c:
apá) ts (ág tS:
becsüle) t t
csalódo) tts (ág tS:
cson) t t
cson) tj c
ezüs) t t
éle) t t
felej) ts tS
gyapo) t t
gráni) t t
já) tsz (ani ts:
tsz ts:
szorí) ts (a tS:
@@ -350,6 +380,8 @@ vörösmar) ty ti
á) t (sző t
á) t (szú t
á) t (szű t
á) t (jut t
á) t (jö t
bizo) tts (ág tS:
állí) ts tS:
állapo) t (sor t
@@ -361,12 +393,17 @@ kiál) ts tS
köve) ts (ég tS:
kür) t t
kür) tj c
küldö) ttj c:
korlátozo) tts tS:
neve) ts (ég tS:
on) ts (u tS
o) tt t:
szen)t (szék t
szé) t t
szöve) ts (ég tS:
söté) ts (ég tS:
sajá) ts (ág tS:
pillan) ts tS
romlo) tts tS:
tar) ts tS
@@ -433,6 +470,7 @@ csontvá) z z
e) z (t s
ho) z (ta s
ho) z (tá s
helye) z (ked s
inté) z (ked s
költö) z (köd s
dictsource/ta_list
+68 -2
View File
@@ -1,4 +1,3 @@
//_xx விழுக்காடு $text // TESTING doesn't reduce vowels
// This file is UTF8 encoded
// Spelling to phoneme words and exceptions for Tamil
@@ -157,6 +156,73 @@ _1M4 a:jirVmko:d.ie:
_0M5 a:jirVma:jirVmko:d.i // 10,000,000,000,000
_1M5 a:jirVma:jirVmko:d.i
_dpt _pul.l.i
// exceptions
பாபு $alt // initial ப as [b]
பாலு $alt
பவானி $alt
பகவதி $alt
பானம் $alt
பாறம் $alt
பாரதி $alt
பாரதம் $alt
பரதம் $alt
பாப $alt
பாகம் $alt
பண்தம் $alt
பவனி $alt
தானம் $alt // initial த as [d]
தேகம் $alt
திரவியம் $alt
தூபம் $alt
தீபம் $alt
தீபாவளி $alt
தீரம் $alt
தைரியம் $alt
திடம் $alt
திரவம் $alt
திடீர் $alt
திவசம் $alt
தியானம் $alt
தானியம் $alt
தண்டனை $alt
துவம்சம் $alt
திருஷ்டி $alt
தயாநிதி $alt
தயாளு $alt
துக்கம் $alt
திகில் $alt
துந்துவி $alt
தயவு $alt
தடால் $alt
திக் $alt
தடக் $alt
தீக்ஷை $alt
கக $alt // initial க as [g]
கனம் $alt
கணம் $alt
கணபதி $alt
கணீர் $alt
கிலி $alt
கில்லி $alt
கோதுமை $alt
கோபுரம் $alt
காந்தம் $alt
காந்தி $alt
குண்டு $alt
கோலி $alt
குகை $alt
கிரி $alt
கோமதி $alt
பப pVpV
dictsource/ta_rules
+44 -1
View File
@@ -43,7 +43,7 @@
_) எ ;e // add a short [j] sound at start of word ?
ஏ e:
_) ஏ ;e:: // add a short [j] sound at start of word ?
_) ஏ ;e: // add a short [j] sound at start of word ?
ஐ aI
@@ -62,12 +62,22 @@
்) க (B g
_) க kV
_) க (B k
_) க (T gV // if word has $alt attribute in ta_list
_) க (BT g // $alt
க்க kkV
க்க (B kk
_) க (ௌ g
க (ீத g
க (ஜ gV
_) க (ம்பீர gV
_) க (ந்தர்வ gV
_) க (ோவிந்த g
ங NV
ங (B N
ச sV // ?? [z]
ச (B s
_) ச sa
@@ -81,12 +91,15 @@
ஞ்) ச dZV
ஞ்) ச (B dZ
ஜ dZV
ஜ (B dZ
ஞ n^V
ஞ (B n^
ட d.V
ட (B d.
_) ட t.V
@@ -94,26 +107,43 @@
ட்ட t.t.V
ட்ட (B t.t.
ண n.V
ண (B n.
த dV
த (B d
_) த tV
_) த (B t
_) த (T dV // if word has $alt attribute in ta_list
_) த (BT d // if word has $alt
த்த ttV
த்த (B tt
// exceptions, த as [d]
_) த (ன dV
_) த (ர்ம dV
_) த (ுர் d
_) த (ூர d
_) த (ெய்வ d
_) த (ுஷ் d
_) த (க்ஷிணா dV
ந nV
ந (B n // dental n
ன nV
ன (B n // alveolar n
ப bV
ப (B b
_) ப pV
_) ப (B p
_) ப (T bV // if word has $alt attribute in ta_list
_) ப (BT b // if word has $alt
ப்ப ppV
ப்ப (B pp
ட்) ப pV
@@ -123,43 +153,56 @@
ஃ) ப fV
ஃ) ப (B f
ம mV
ம (B m
ய jV
ய (B j
ர rV
ர (B r
ற RV
ற (B R
ற் (ற t. // RR -> t.R
ல lV
ல (B l
ள l.V
ள (B l.
ழ z.V
ழ (B z.
வ vV
வ (B v
ஶ SV
ஶ (B S
ஷ s.V
ஷ (B s.
ஸ sV
ஸ (B s
ஹ hV
ஹ (B h
க்ஷ ks.V
க்ஷ (B ks.
phsource/compile_report
+16 -9
View File
@@ -1,4 +1,4 @@
61 phoneme tables
62 phoneme tables
new total
base 104 104
base2 26 125
@@ -61,6 +61,7 @@
eu 6 126
mn 15 115
pa 13 153
prs 8 111
Data file Used by
b/b [b] base
@@ -833,7 +834,6 @@ ustop/kr [k] base
[k] base2
[k] en
[k] fi
[k] fr
[kh] hi
[k] hu
[k] lv
@@ -1185,6 +1185,7 @@ vdiph/aoo [aU] en_us
vdiph/au [aU] en
[aU] af
[a:U] vi
[aU] prs
vdiph/au# [aU] en_sc
vdiph/au_2 [au] zhy
vdiph/au_3 [aU] en_rp
@@ -1204,6 +1205,7 @@ vdiph/eei [EI] base2
[EI] vi
[eI] hy
[eI] ne
[eI] prs
vdiph/eei_2 [eI] eo
[ei] fi
[ei] lv
@@ -1339,11 +1341,11 @@ vnasal/e_n [e~] af
[e~] hi
[e~] pt
[e~] bn
vnasal/ii_n [I~] hi
vnasal/i_n [i~] pt
[i~] bn
[i:~] bn
vnasal/i_n2 [I~] hi
[i~] hi
vnasal/i_n2 [i~] hi
vnasal/m- [m-] sw
vnasal/n- [n-] sw
vnasal/nn- [N-] sw
@@ -1629,6 +1631,7 @@ vowel/aa_2 [A:] en
[A] zh
vowel/aa_3 [A:] af
[A] sq
[A] prs
vowel/aa_4 [A:] sv
[O] vi
vowel/aa_5 [A:] en_n
@@ -1755,6 +1758,7 @@ vowel/e_mid [E] en_rp
[E2] id
[e] hy
[E] bn
[E] prs
vowel/e_mid2 [E] af
[E] de
[E2] de
@@ -1803,6 +1807,7 @@ vowel/i [i] base2
[i:] kn
[i] ne
[i] mn
[i] prs
vowel/i# [y:] cy
vowel/i_2 [i2] de
[i:] de
@@ -1925,6 +1930,7 @@ vowel/o [o] base2
[oU] en_n
[oU] en_wi
[o:] de
[o2] fr
[o] hi
[o] ta
[o] it
@@ -1945,8 +1951,8 @@ vowel/o [o] base2
[o] bn
[o] mr
[o:] mn
[o:] pa
vowel/o_2 [o:] cy
[o2] fr
[o:] hi
[o:] hu
[o:] la
@@ -1954,11 +1960,9 @@ vowel/o_2 [o:] cy
[o] vi
[o] da
[o:] bn
[o:] pa
vowel/o-_2 [V] en_n
[V] en_wm
vowel/o_3 [oU] en_sc
[o] fr
[o:] ta
vowel/o-_3 [U] en_rp
vowel/o-_4 [o] ro
@@ -1966,6 +1970,7 @@ vowel/o_5 [O] nl
[O] da
vowel/o_6 [O:] nl
vowel/o_7 [o] ku
vowel/o_8 [o] fr
vowel/oe [W] en
[W] af
[W] de
@@ -1996,6 +2001,7 @@ vowel/oo [O:] en_sc
[o] zhy
[O] tr
[O] id
[o] prs
vowel/oo_1 [O:] en_n
[O:] en_rp
[O@] en_rp
@@ -2064,6 +2070,7 @@ vowel/u [u:] en_wi
[u] kn
[u:] kn
[u] bn
[u] prs
vowel/u# [u:] en_sc
[Y] tr
vowel/u_2 [u1] fi
@@ -2250,8 +2257,7 @@ vwl_en_us/oor [0] en_us
vwl_en_us/or [o@] en_us
[O:] en_sc
vwl_en_us/ur [U@] en_us
vwl_fr/r [R] fr
[r] fr
vwl_fr/r [r] fr
[r2] fr
vwl_fr/_r [R] fr
[r] fr
@@ -2265,6 +2271,7 @@ vwl_fr/r@ [@] fr
[E~] fr
[W~] fr
vwl_fr/@R [x] pt_pt
vwl_fr/_r2 [R] fr
vwl_fr/@R2 [R] fr_ca
vwl_fr/@R3 [Q2] hy
vwl_fr/@R4 [Q2] hy
phsource/ph_french
+11 -11
View File
@@ -1,7 +1,7 @@
//====================================================
// French
//====================================================
// Updated 2009-09-30 Michel Such <michel.such@free.fr>
// Updated 2009-10-07 Michel Such <michel.such@free.fr>
phoneme (l) virtual
// Used for l and l/
@@ -96,7 +96,7 @@ endphoneme
phoneme i- // shorter
vowel starttype (i) endtype (i)
length 130
length 110
formants vowel/i_2
after (l) l/li-15
after (r) vwl_fr/ri
@@ -119,7 +119,7 @@ endphoneme
phoneme o
vowel starttype (o) endtype (o)
length 170
formants vowel/o_3
formants vowel/o_8
after (l) l/lo-10
after (r) vwl_fr/ro
before l/ DFT+l/l_o
@@ -132,7 +132,7 @@ endphoneme
phoneme o2
vowel starttype (o) endtype (o)
length 170
formants vowel/o_2
formants vowel/o
after (l) l/lo-10
after (r) vwl_fr/ro
before l/ vowel/o_mid2+l/l_o
@@ -237,10 +237,10 @@ endphoneme
phoneme wA
vowel starttype (o) endtype (a)
length 230
length 220
formants vwl_fr/w_a
after (l) l/lo-10
after (r) vwl_fr/ro-15
after (r) vwl_fr/ro
before l/ DFT-10+l/l_a
before j DFT
before (r) DFT+vwl_fr/xr
@@ -394,7 +394,7 @@ phoneme k
lengthmod 2
wave ustop/k%40 // reduce strength of noise burst
before _ ustop/k_
before (r) ustop/kr%45
before (r) ustop/k%50
before r/2 ustop/k%60
before (l) ustop/kl%45
before (i) ustop/ki%45
@@ -465,7 +465,7 @@ endphoneme
phoneme r // between vowels in a word
liquid uvl starttype (r) endtype (r)
vowelin f1=0 f2=1600 -300 300 f3=-200 80
vowelin f1=0 f2=1600 -300 300 f3=-200 80 len=20
vowelout f1=2 f2=1600 -300 300 f3=-300 80
length 100
lengthmod 7
@@ -481,12 +481,12 @@ endphoneme
phoneme R // First letter of a word
liquid uvl starttype (r) endtype (r)
vowelin f1=0 f2=1600 -300 300 f3=-200 80
vowelin f1=0 f2=1600 -300 300 f3=-200 80 len=20
vowelout f1=2 f2=1600 -300 300 f3=-300 80
length 100
lengthmod 0
beforenotvowel r/2
formants vwl_fr/r
formants vwl_fr/_r2
after _ vwl_fr/_r
after (a) vwl_fr/_r
after (e) vwl_fr/_r
@@ -513,7 +513,7 @@ endphoneme
phoneme r/2 // variant of [r] when not preceding a vowel
liquid uvl starttype r/2 endtype r/2
vowelout f1=2 f2=1600 -300 300 f3=-300 80
vowelout f1=2 f2=1600 -300 300 f3=-300 100
length 100
lengthmod 2
formants vwl_fr/r_+r3/rx%25
phsource/ph_hindi
+1 -1
View File
@@ -169,7 +169,7 @@ endphoneme
phoneme I~
vowel starttype (i) endtype (i)
length 170
formants vnasal/i_n2
formants vnasal/ii_n
endphoneme
phoneme e~
phsource/phonemes
+3
View File
@@ -1382,3 +1382,6 @@ include ph_mongolian
phonemetable pa hi
include ph_punjabi
phonemetable prs base
include ph_dari
phsource/vnasal/aa_n4
Executable
BIN
View File
Binary file not shown.
phsource/vnasal/i_n2
BIN
View File
Binary file not shown.
phsource/vnasal/ii_n
BIN
View File
Binary file not shown.
phsource/vowel/V_6
BIN
View File
Binary file not shown.
phsource/vwl_fr/w_a
BIN
View File
Binary file not shown.
src/compiledict.cpp
+1 -4
View File
@@ -223,8 +223,6 @@ static int compile_line(char *linebuf, char *dict_line, int *hash)
unsigned char bad_phoneme[4];
static char nullstring[] = {0};
WORD_TAB winfo;
comment = NULL;
text_not_phonemes = 0;
phonetic = word = nullstring;
@@ -437,12 +435,11 @@ step=1; // TEST
if(word[0] == '_')
{
// This is a special word, used by eSpeak. Translate this into phonemes now
memset(&winfo,0,sizeof(winfo));
strcat(phonetic, " "); // need a space to indicate word-boundary
// PROBLEM vowel reductions are not applied to the translated phonemes
// condition rules are not applied
TranslateWord(translator,phonetic,0,&winfo);
TranslateWord(translator,phonetic,0,NULL);
text_not_phonemes = 0;
strncpy0(encoded_ph, word_phonemes, N_WORD_BYTES-4);
src/dictionary.cpp
+6 -3
View File
@@ -247,6 +247,12 @@ int LoadDictionary(Translator *tr, const char *name, int no_error)
sprintf(fname,"%s%c%s_dict",path_home,PATHSEP,name);
size = GetFileLength(fname);
if(tr->data_dictlist != NULL)
{
Free(tr->data_dictlist);
tr->data_dictlist = NULL;
}
f = fopen(fname,"rb");
if((f == NULL) || (size <= 0))
{
@@ -257,9 +263,6 @@ int LoadDictionary(Translator *tr, const char *name, int no_error)
return(1);
}
if(tr->data_dictlist != NULL)
Free(tr->data_dictlist);
tr->data_dictlist = Alloc(size);
fread(tr->data_dictlist,size,1,f);
fclose(f);
src/extras.cpp
+4 -11
View File
@@ -317,7 +317,6 @@ void Lexicon_It(int pass)
char phonemes[80];
char phonemes2[80];
char buf_out[120];
WORD_TAB winfo;
static const char *vowels1 = "aeiou";
static const char *vowels2 = "aeou";
@@ -438,8 +437,7 @@ void Lexicon_It(int pass)
}
}
// translate
memset(&winfo,0,sizeof(winfo));
TranslateWord(translator,&word1[1],0,&winfo);
TranslateWord(translator,&word1[1],0, NULL);
DecodePhonemes(word_phonemes,phonemes);
stress_posn1 = 0;
@@ -462,8 +460,7 @@ void Lexicon_It(int pass)
vowel_ix++;
}
memset(&winfo,0,sizeof(winfo));
TranslateWord(translator,&word2[1],0,&winfo);
TranslateWord(translator,&word2[1],0, NULL);
DecodePhonemes(word_phonemes,phonemes2);
vowel_ix = 1;
@@ -619,7 +616,6 @@ void Lexicon_De()
char pronounce2[80];
char phonemes[80];
char phonemes2[80];
WORD_TAB winfo;
static const char *vowels = "aeiouyAEIOUY29@";
@@ -728,8 +724,7 @@ void Lexicon_De()
}
// translate
memset(&winfo,0,sizeof(winfo));
TranslateWord(translator,&word2[1],0,&winfo);
TranslateWord(translator,&word2[1],0, NULL);
DecodePhonemes2(word_phonemes,phonemes); // also need to change some phoneme names
@@ -794,7 +789,6 @@ void Lexicon_Ru()
int wlen;
int len;
int check_root;
WORD_TAB winfo;
char word[80];
char word2[80];
@@ -933,8 +927,7 @@ p_unicode = unicode;
}
// translate
memset(&winfo,0,sizeof(winfo));
TranslateWord(translator, &word2[1],0,&winfo);
TranslateWord(translator, &word2[1],0, NULL);
DecodePhonemes(word_phonemes,phonemes);
// find the stress position in the translation
src/numbers.cpp
+13 -18
View File
@@ -1314,7 +1314,7 @@ static int TranslateNumber_1(Translator *tr, char *word, char *ph_out, unsigned
char ph_buf2[50];
char suffix[20];
char *wordptr;
unsigned int dictflags[2];
unsigned int dictflags;
static const char str_pause[2] = {phonPAUSE_NOLINK,0};
@@ -1331,14 +1331,12 @@ static int TranslateNumber_1(Translator *tr, char *word, char *ph_out, unsigned
// but not if the next word starts with an upper-case letter
ordinal = 2;
word[ix] = ' ';
if(tr->translator_name == L('h','u'))
{
// lang=hu don't treat dot as ordinal indicator if the next word is a month name ($alt)
dictflags[0] = 0;
wordptr = &word[ix+2];
LookupDictList(tr, &wordptr, ph_buf, dictflags, 0, NULL);
if(dictflags[0] & FLAG_ALT_TRANS) // TEST
dictflags = TranslateWord(tr, &word[ix+2], 0, NULL);
if(dictflags & FLAG_ALT_TRANS)
ordinal = 0;
}
}
@@ -1514,23 +1512,20 @@ static int TranslateNumber_1(Translator *tr, char *word, char *ph_out, unsigned
}
break;
case NUM_DFRACTION_1: // italian, say "hundredths" is leading zero
case NUM_DFRACTION_1: // italian, say "hundredths" if leading zero
case NUM_DFRACTION_5: // hungarian, always say "tenths" etc.
if(decimal_count <= 4)
LookupNum3(tr, atoi(&word[n_digits]), ph_buf, 0,0,0);
if((word[n_digits]=='0') || (decimal_mode == NUM_DFRACTION_5))
{
LookupNum3(tr, atoi(&word[n_digits]), ph_buf, 0,0,0);
if((word[n_digits]=='0') || (decimal_mode == 0xa000))
{
// decimal part has leading zeros, so add a "hundredths" or "thousandths" suffix
sprintf(string,"_0Z%d",decimal_count);
if(Lookup(tr, string, buf1) == 0)
break; // revert to speaking single digits
// decimal part has leading zeros, so add a "hundredths" or "thousandths" suffix
sprintf(string,"_0Z%d",decimal_count);
if(Lookup(tr, string, buf1) == 0)
break; // revert to speaking single digits
strcat(ph_buf,buf1);
}
strcat(ph_out,ph_buf);
n_digits += decimal_count;
strcat(ph_buf,buf1);
}
strcat(ph_out,ph_buf);
n_digits += decimal_count;
break;
case NUM_DFRACTION_3:
src/readclause.cpp
+106 -55
View File
@@ -816,8 +816,8 @@ static int LoadSoundFile2(const char *fname)
static int AnnouncePunctuation(Translator *tr, int c1, int c2, char *buf, int bufix)
{//=================================================================================
static int AnnouncePunctuation(Translator *tr, int c1, int *c2_ptr, char *output, int *bufix, int end_clause)
{//==========================================================================================================
// announce punctuation names
// c1: the punctuation character
// c2: the following character
@@ -826,12 +826,19 @@ static int AnnouncePunctuation(Translator *tr, int c1, int c2, char *buf, int bu
const char *punctname;
int found = 0;
int soundicon;
char *p;
int c2;
int len;
int bufix1;
char buf[200];
char buf2[80];
c2 = *c2_ptr;
buf[0] = 0;
if((soundicon = LookupSoundicon(c1)) >= 0)
{
// add an embedded command to play the soundicon
sprintf(&buf[bufix],"\001%dI ",soundicon);
sprintf(buf,"\001%dI ",soundicon);
UngetC(c2);
found = 1;
}
@@ -839,7 +846,7 @@ static int AnnouncePunctuation(Translator *tr, int c1, int c2, char *buf, int bu
if((punctname = LookupCharName(tr, c1)) != NULL)
{
found = 1;
if(bufix==0)
if((*bufix==0) || (end_clause==0))
{
punct_count=1;
while(c2 == c1)
@@ -847,26 +854,32 @@ static int AnnouncePunctuation(Translator *tr, int c1, int c2, char *buf, int bu
punct_count++;
c2 = GetC();
}
UngetC(c2);
*c2_ptr = c2;
if(end_clause)
{
UngetC(c2);
}
p = &buf[bufix];
if(punct_count==1)
{
sprintf(p,"%s %s %s",tone_punct_on,punctname,tone_punct_off);
sprintf(buf,"%s %s %s",tone_punct_on,punctname,tone_punct_off);
}
else
if(punct_count < 4)
{
sprintf(p,"\001+10S%s",tone_punct_on);
sprintf(buf,"\001+10S%s",tone_punct_on);
while(punct_count-- > 0)
sprintf(buf,"%s %s",buf,punctname);
sprintf(p,"%s %s\001-10S",buf,tone_punct_off);
{
sprintf(buf2," %s",punctname);
strcat(buf, buf2);
}
sprintf(buf2," %s\001-10S",tone_punct_off);
strcat(buf, buf2);
}
else
{
sprintf(p,"%s %s %d %s %s",
sprintf(buf,"%s %s %d %s %s",
tone_punct_on,punctname,punct_count,punctname,tone_punct_off);
return(CLAUSE_COMMA);
}
}
else
@@ -879,17 +892,25 @@ static int AnnouncePunctuation(Translator *tr, int c1, int c2, char *buf, int bu
ssml_ignore_l_angle = c1; // this was &lt; which was converted to <, don't pick it up again as <
}
ungot_char2 = c1;
buf[bufix] = ' ';
buf[bufix+1] = 0;
buf[0] = ' ';
buf[1] = 0;
}
}
if(found == 0)
return(-1);
bufix1 = *bufix;
len = strlen(buf);
strcpy(&output[*bufix],buf);
*bufix += len;
if(end_clause==0)
return(-1);
if(c1 == '-')
return(CLAUSE_NONE); // no pause
if(bufix > 0)
if(bufix1 > 0)
return(CLAUSE_SHORTCOMMA);
if((strchr_w(punct_close,c1) != NULL) && !iswalnum(c2))
return(CLAUSE_SHORTFALL+4);
@@ -1919,8 +1940,8 @@ int ReadClause(Translator *tr, FILE *f_in, char *buf, short *charix, int *charix
int any_alnum = 0;
int self_closing;
int punct_data;
int is_punctuation;
int save_c2;
int is_end_clause;
int announced_punctuation;
int stressed_word = 0;
const char *p;
wchar_t xml_buf[N_XML_BUF+1];
@@ -2304,35 +2325,48 @@ if(option_ssml) parag=1;
linelength = 0;
}
if(option_punctuation && (phoneme_mode==0) && (sayas_mode==0) && iswpunct(c1))
{
// option is set to explicitly speak punctuation characters
// if a list of allowed punctuation has been set up, check whether the character is in it
if((option_punctuation == 1) || (wcschr(option_punctlist,c1) != NULL))
{
if((terminator = AnnouncePunctuation(tr, c1, c2, buf, ix)) >= 0)
return(terminator);
}
}
announced_punctuation = 0;
if((phoneme_mode==0) && (sayas_mode==0) && ((punct = lookupwchar(punct_chars,c1)) != 0))
if((phoneme_mode==0) && (sayas_mode==0))
{
punct_data = punct_attributes[punct];
is_end_clause = 0;
if(punct_data & PUNCT_IN_WORD)
if((punct = lookupwchar(punct_chars,c1)) != 0)
{
// Armenian punctuation inside a word
stressed_word = 1;
*tone_type = punct_data >> 12 & 0xf; // override the end-of-sentence type
continue;
punct_data = punct_attributes[punct];
if(punct_data & PUNCT_IN_WORD)
{
// Armenian punctuation inside a word
stressed_word = 1;
*tone_type = punct_data >> 12 & 0xf; // override the end-of-sentence type
continue;
}
if((iswspace(c2) || (punct_data & 0x8000) || IsBracket(c2) || (c2=='?') || (c2=='-') || Eof()))
{
// note: (c2='?') is for when a smart-quote has been replaced by '?'
is_end_clause = 1;
}
}
if((iswspace(c2) || (punct_data & 0x8000) || IsBracket(c2) || (c2=='?') || (c2=='-') || Eof()))
if(option_punctuation && iswpunct(c1))
{
// note: (c2='?') is for when a smart-quote has been replaced by '?'
is_punctuation = 1;
// option is set to explicitly speak punctuation characters
// if a list of allowed punctuation has been set up, check whether the character is in it
if((option_punctuation == 1) || (wcschr(option_punctlist,c1) != NULL))
{
if((terminator = AnnouncePunctuation(tr, c1, &c2, buf, &ix, is_end_clause)) >= 0)
return(terminator);
announced_punctuation = c1;
}
}
if(is_end_clause)
{
int c_next;
if((c1 == '.') && (cprev == '.'))
{
c1 = 0x2026;
@@ -2340,47 +2374,43 @@ if(option_ssml) parag=1;
}
nl_count = 0;
save_c2 = c2;
c_next = c2;
if(iswspace(c2))
if(iswspace(c_next))
{
while(!Eof() && iswspace(c2))
while(!Eof() && iswspace(c_next))
{
if(c2 == '\n')
if(c_next == '\n')
nl_count++;
c2 = GetC(); // skip past space(s)
}
if(!Eof())
{
UngetC(c2);
c_next = GetC(); // skip past space(s)
}
}
if((nl_count==0) && (c1 == '.'))
{
// if(iswdigit(cprev) && (tr->langopts.numbers & NUM_ORDINAL_DOT) && islower(c2))
if(iswdigit(cprev) && (tr->langopts.numbers & NUM_ORDINAL_DOT))
{
// dot after a number indicates an ordinal number
is_punctuation = 0;
if(islower(c_next) || (c_next == '<'))
is_end_clause = 0; // only if followed by lower-case, (or if there is a XML tag)
}
else
if(iswlower(c2))
if(iswlower(c_next))
{
// next word has no capital letter, this dot is probably from an abbreviation
c1 = ' ';
is_punctuation = 0;
is_end_clause = 0;
}
if(any_alnum==0)
{
c1 = ' '; // no letters or digits yet, so probably not a sentence terminator
is_punctuation = 0;
is_end_clause = 0;
}
}
c2 = save_c2;
if(is_punctuation)
if(is_end_clause)
{
UngetC(c_next);
buf[ix] = ' ';
buf[ix+1] = 0;
@@ -2393,12 +2423,33 @@ if(option_ssml) parag=1;
}
return(punct_data); // only recognise punctuation if followed by a blank or bracket/quote
}
else
{
if(!Eof())
{
if(iswspace(c2))
UngetC(c_next);
}
}
}
}
if(speech_parameters[espeakSILENCE]==1)
continue;
if(c1 == announced_punctuation)
{
// const unsigned short keep_punctuation[] = {
// '\'', '-', 0x92, 0xb4, 0x2019, 0x2032, 0 };
// This character has already been announced, so delete it so that it isn't spoken a second time.
// Unless it's a hyphen or apostrophe (which is used by TranslateClause() )
if(!IsBracket(c1))
{
c1 = ' ';
}
}
j = ix+1;
ix += utf8_out(c1,&buf[ix]); // buf[ix++] = c1;
if(!iswspace(c1) && !IsBracket(c1))
src/synthdata.cpp
+1 -1
View File
@@ -35,7 +35,7 @@
#include "translate.h"
#include "wave.h"
const char *version_string = "1.41.13 10.Oct.09";
const char *version_string = "1.41.15 12.Oct.09";
const int version_phdata = 0x014100;
int option_device_number = -1;
src/synthesize.cpp
+1 -1
View File
@@ -535,7 +535,7 @@ static short vcolouring[N_VCOLOUR][5] = {
// fprintf(stderr,"FMT%d %3s %3d-%3d f1=%d f2=%4d %4d %4d f3=%4d %3d\n",
// which,WordToString(other_ph->mnemonic),len,rms,f1,f2,f2_min,f2_max,f3_adj,f3_amp);
if(other_ph->mnemonic == '?')
if((other_ph != NULL) && (other_ph->mnemonic == '?'))
flags |= 8;
if(which == 1)
src/tr_languages.cpp
+7
View File
@@ -464,6 +464,13 @@ Translator *SelectTranslator(const char *name)
break;
case L('f','a'): // Farsi
{
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 1; // disable check for unpronouncable words
}
break;
case L('f','i'): // Finnish
{
static const unsigned char stress_amps_fi[8] = {18,16, 22,22, 20,22, 22,22 };
src/translate.cpp
+19 -2
View File
@@ -601,8 +601,9 @@ int TranslateWord(Translator *tr, char *word1, int next_pause, WORD_TAB *wtab)
int spell_word;
int stress_bits;
int emphasize_allcaps = 0;
int wflags = wtab->flags;
int wmark = wtab->wmark;
int wflags;
int wmark;
WORD_TAB wtab_null[8];
// translate these to get pronunciations of plural 's' suffix (different forms depending on
// the preceding letter
@@ -610,6 +611,14 @@ int TranslateWord(Translator *tr, char *word1, int next_pause, WORD_TAB *wtab)
static char word_iz[4] = {0,'i','z',0};
static char word_ss[4] = {0,'s','s',0};
if(wtab == NULL)
{
memset(wtab_null, 0, sizeof(wtab_null));
wtab = wtab_null;
}
wflags = wtab->flags;
wmark = wtab->wmark;
dictionary_flags[0] = 0;
dictionary_flags[1] = 0;
dictionary_flags2[0] = 0;
@@ -620,6 +629,13 @@ int TranslateWord(Translator *tr, char *word1, int next_pause, WORD_TAB *wtab)
end_phonemes[0] = 0;
ph_limit = &phonemes[N_WORD_PHONEMES];
if(tr->data_dictlist == NULL)
{
// dictionary is not loaded
word_phonemes[0] = 0;
return(0);
}
// count the length of the word
if(*word1 == ' ') word1++; // possibly a dot was replaced by space: $dot
wordx = word1;
@@ -1053,6 +1069,7 @@ strcpy(phonemes2,phonemes);
AppendPhonemes(tr,phonemes, N_WORD_PHONEMES, end_phonemes);
end_phonemes[0] = 0;
}
memcpy(wordx,word_copy,strlen(word_copy));
}
wordx[-1] = c_temp;
}
src/translate.h
+1 -1
View File
@@ -98,10 +98,10 @@
#define FLAG_DONT_SWITCH_TRANSLATOR 0x1000
#define FLAG_SUFFIX_REMOVED 0x2000
#define FLAG_HYPHEN_AFTER 0x4000
#define FLAG_ORDINAL 0x8000 // passed to TranslateNumber() to indicate an ordinal number
#define FLAG_NO_TRACE 0x10000 // passed to TranslateRules() to suppress dictionary lookup printout
#define FLAG_NO_PREFIX 0x20000
#define FLAG_ORDINAL 0x40000 // passed to TranslateNumber() to indicate an ordinal number
// prefix/suffix flags (bits 8 to 14, bits 16 to 22) don't use 0x8000, 0x800000
#define SUFX_E 0x0100 // e may have been added