forked from espeak-ng/espeak-ng
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtr_languages.c
1681 lines (1491 loc) · 71.1 KB
/
tr_languages.c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/*
* Copyright (C) 2005 to 2015 by Jonathan Duddington
* email: [email protected]
* Copyright (C) 2015-2016, 2020 Reece H. Dunn
* Copyright (C) 2021 Juho Hiltunen
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, see: <http://www.gnu.org/licenses/>.
*/
#include "config.h"
#include <ctype.h>
#include <locale.h>
#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <espeak-ng/espeak_ng.h>
#include <espeak-ng/speak_lib.h>
#include <espeak-ng/encoding.h>
#include "common.h"
#include "setlengths.h" // for SetLengthMods
#include "translate.h" // for Translator, LANGUAGE_OPTIONS, L, NUM...
// start of unicode pages for character sets
#define OFFSET_GREEK 0x380
#define OFFSET_CYRILLIC 0x420
#define OFFSET_ARMENIAN 0x530
#define OFFSET_HEBREW 0x590
#define OFFSET_ARABIC 0x600
#define OFFSET_SYRIAC 0x700
#define OFFSET_DEVANAGARI 0x900
#define OFFSET_BENGALI 0x980
#define OFFSET_GURMUKHI 0xa00
#define OFFSET_GUJARATI 0xa80
#define OFFSET_ORIYA 0xb00
#define OFFSET_TAMIL 0xb80
#define OFFSET_TELUGU 0xc00
#define OFFSET_KANNADA 0xc80
#define OFFSET_MALAYALAM 0xd00
#define OFFSET_SINHALA 0x0d80
#define OFFSET_THAI 0x0e00
#define OFFSET_LAO 0x0e80
#define OFFSET_TIBET 0x0f00
#define OFFSET_MYANMAR 0x1000
#define OFFSET_GEORGIAN 0x10a0
#define OFFSET_KOREAN 0x1100
#define OFFSET_ETHIOPIC 0x1200
// character ranges must be listed in ascending unicode order
static const ALPHABET alphabets[] = {
{ "_el", OFFSET_GREEK, 0x380, 0x3ff, L('e', 'l'), AL_DONT_NAME | AL_NOT_LETTERS | AL_WORDS },
{ "_cyr", OFFSET_CYRILLIC, 0x400, 0x52f, 0, 0 },
{ "_hy", OFFSET_ARMENIAN, 0x530, 0x58f, L('h', 'y'), AL_WORDS },
{ "_he", OFFSET_HEBREW, 0x590, 0x5ff, 0, 0 },
{ "_ar", OFFSET_ARABIC, 0x600, 0x6ff, 0, 0 },
{ "_syc", OFFSET_SYRIAC, 0x700, 0x74f, 0, 0 },
{ "_hi", OFFSET_DEVANAGARI, 0x900, 0x97f, L('h', 'i'), AL_WORDS },
{ "_bn", OFFSET_BENGALI, 0x0980, 0x9ff, L('b', 'n'), AL_WORDS },
{ "_gur", OFFSET_GURMUKHI, 0xa00, 0xa7f, L('p', 'a'), AL_WORDS },
{ "_gu", OFFSET_GUJARATI, 0xa80, 0xaff, L('g', 'u'), AL_WORDS },
{ "_or", OFFSET_ORIYA, 0xb00, 0xb7f, 0, 0 },
{ "_ta", OFFSET_TAMIL, 0xb80, 0xbff, L('t', 'a'), AL_WORDS },
{ "_te", OFFSET_TELUGU, 0xc00, 0xc7f, L('t', 'e'), 0 },
{ "_kn", OFFSET_KANNADA, 0xc80, 0xcff, L('k', 'n'), AL_WORDS },
{ "_ml", OFFSET_MALAYALAM, 0xd00, 0xd7f, L('m', 'l'), AL_WORDS },
{ "_si", OFFSET_SINHALA, 0xd80, 0xdff, L('s', 'i'), AL_WORDS },
{ "_th", OFFSET_THAI, 0xe00, 0xe7f, 0, 0 },
{ "_lo", OFFSET_LAO, 0xe80, 0xeff, 0, 0 },
{ "_ti", OFFSET_TIBET, 0xf00, 0xfff, 0, 0 },
{ "_my", OFFSET_MYANMAR, 0x1000, 0x109f, 0, 0 },
{ "_ka", OFFSET_GEORGIAN, 0x10a0, 0x10ff, L('k', 'a'), AL_WORDS },
{ "_ko", OFFSET_KOREAN, 0x1100, 0x11ff, L('k', 'o'), AL_WORDS },
{ "_eth", OFFSET_ETHIOPIC, 0x1200, 0x139f, 0, 0 },
{ "_braille", 0x2800, 0x2800, 0x28ff, 0, AL_NO_SYMBOL },
{ "_ja", 0x3040, 0x3040, 0x30ff, 0, AL_NOT_CODE },
{ "_zh", 0x3100, 0x3100, 0x9fff, 0, AL_NOT_CODE },
{ "_ko", 0xa700, 0xa700, 0xd7ff, L('k', 'o'), AL_NOT_CODE | AL_WORDS },
{ NULL, 0, 0, 0, 0, 0 }
};
const ALPHABET *AlphabetFromChar(int c)
{
// Find the alphabet from a character.
const ALPHABET *alphabet = alphabets;
while (alphabet->name != NULL) {
if (c <= alphabet->range_max) {
if (c >= alphabet->range_min)
return alphabet;
else
break;
}
alphabet++;
}
return NULL;
}
static void Translator_Russian(Translator *tr);
static void SetLetterVowel(Translator *tr, int c)
{
tr->letter_bits[c] = (tr->letter_bits[c] & 0x40) | 0x81; // keep value for group 6 (front vowels e,i,y)
}
static void ResetLetterBits(Translator *tr, int groups)
{
// Clear all the specified groups
unsigned int ix;
unsigned int mask;
mask = ~groups;
for (ix = 0; ix < sizeof(tr->letter_bits); ix++)
tr->letter_bits[ix] &= mask;
}
static void SetLetterBits(Translator *tr, int group, const char *string)
{
int bits;
unsigned char c;
bits = (1L << group);
while ((c = *string++) != 0)
tr->letter_bits[c] |= bits;
}
static void SetLetterBitsRange(Translator *tr, int group, int first, int last)
{
int bits;
int ix;
bits = (1L << group);
for (ix = first; ix <= last; ix++)
tr->letter_bits[ix] |= bits;
}
static void SetLetterBitsUTF8(Translator *tr, int group, const char *letters, int offset)
{
// Add the letters to the specified letter group.
const char *p = letters;
int code = -1;
while (code != 0) {
int bytes = utf8_in(&code, p);
if (code > 0x20)
tr->letter_bits[code - offset] |= (1L << group);
p += bytes;
}
}
// ignore these characters
static const unsigned short chars_ignore_default[] = {
// U+00AD SOFT HYPHEN
// Used to mark hyphenation points in words for where to split a
// word at the end of a line to provide readable justified text.
0xad, 1,
// U+200C ZERO WIDTH NON-JOINER
// Used to prevent combined ligatures being displayed in their
// combined form.
0x200c, 1,
// U+200D ZERO WIDTH JOINER
// Used to indicate an alternative connected form made up of the
// characters surrounding the ZWJ in Devanagari, Kannada, Malayalam
// and Emoji.
// 0x200d, 1, // Not ignored.
// End of the ignored character list.
0, 0
};
// alternatively, ignore characters but allow zero-width-non-joiner (lang-fa)
static const unsigned short chars_ignore_zwnj_hyphen[] = {
// U+00AD SOFT HYPHEN
// Used to mark hyphenation points in words for where to split a
// word at the end of a line to provide readable justified text.
0xad, 1,
// U+0640 TATWEEL (KASHIDA)
// Used in Arabic scripts to stretch characters for justifying
// the text.
0x640, 1,
// U+200C ZERO WIDTH NON-JOINER
// Used to prevent combined ligatures being displayed in their
// combined form.
0x200c, '-',
// U+200D ZERO WIDTH JOINER
// Used to indicate an alternative connected form made up of the
// characters surrounding the ZWJ in Devanagari, Kannada, Malayalam
// and Emoji.
// 0x200d, 1, // Not ignored.
// End of the ignored character list.
0, 0
};
static const unsigned char utf8_ordinal[] = { 0xc2, 0xba, 0 }; // masculine ordinal character, UTF-8
static const unsigned char utf8_null[] = { 0 }; // null string, UTF-8
static Translator *NewTranslator(void)
{
Translator *tr;
int ix;
static const unsigned char stress_amps2[] = { 18, 18, 20, 20, 20, 22, 22, 20 };
static const short stress_lengths2[8] = { 182, 140, 220, 220, 220, 240, 260, 280 };
static const wchar_t empty_wstring[1] = { 0 };
static const wchar_t punct_in_word[2] = { '\'', 0 }; // allow hyphen within words
static const unsigned char default_tunes[6] = { 0, 1, 2, 3, 0, 0 };
// Translates character codes in the range transpose_min to transpose_max to
// a number in the range 1 to 63. 0 indicates there is no translation.
// Used up to 57 (max of 63)
static const char transpose_map_latin[] = {
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, // 0x60
16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 0, 0, 0, 0, 0, // 0x70
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x80
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x90
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0xa0
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0xb0
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0xc0
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0xd0
27, 28, 29, 0, 0, 30, 31, 32, 33, 34, 35, 36, 0, 37, 38, 0, // 0xe0
0, 0, 0, 39, 0, 0, 40, 0, 41, 0, 42, 0, 43, 0, 0, 0, // 0xf0
0, 0, 0, 44, 0, 45, 0, 46, 0, 0, 0, 0, 0, 47, 0, 0, // 0x100
0, 48, 0, 0, 0, 0, 0, 0, 0, 49, 0, 0, 0, 0, 0, 0, // 0x110
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x120
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x130
0, 0, 50, 0, 51, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x140
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 52, 0, 0, 0, 0, // 0x150
0, 53, 0, 54, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x160
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 55, 0, 56, 0, 57, 0, // 0x170
};
if ((tr = (Translator *)malloc(sizeof(Translator))) == NULL)
return NULL;
tr->encoding = ESPEAKNG_ENCODING_ISO_8859_1;
dictionary_name[0] = 0;
tr->dictionary_name[0] = 0;
tr->phonemes_repeat[0] = 0;
tr->dict_condition = 0;
tr->dict_min_size = 0;
tr->data_dictrules = NULL; // language_1 translation rules file
tr->data_dictlist = NULL; // language_2 dictionary lookup file
tr->transpose_min = 0x60;
tr->transpose_max = 0x17f;
tr->transpose_map = transpose_map_latin;
tr->frequent_pairs = NULL;
tr->expect_verb = 0;
tr->expect_past = 0;
tr->expect_verb_s = 0;
tr->expect_noun = 0;
tr->clause_upper_count = 0;
tr->clause_lower_count = 0;
// only need lower case
tr->letter_bits_offset = 0;
memset(tr->letter_bits, 0, sizeof(tr->letter_bits));
memset(tr->letter_groups, 0, sizeof(tr->letter_groups));
// 0-6 sets of characters matched by A B C H F G Y in pronunciation rules
// these may be set differently for different languages
SetLetterBits(tr, 0, "aeiou"); // A vowels, except y
SetLetterBits(tr, 1, "bcdfgjklmnpqstvxz"); // B hard consonants, excluding h,r,w
SetLetterBits(tr, 2, "bcdfghjklmnpqrstvwxz"); // C all consonants
SetLetterBits(tr, 3, "hlmnr"); // H 'soft' consonants
SetLetterBits(tr, 4, "cfhkpqstx"); // F voiceless consonants
SetLetterBits(tr, 5, "bdgjlmnrvwyz"); // G voiced
SetLetterBits(tr, 6, "eiy"); // Letter group Y, front vowels
SetLetterBits(tr, 7, "aeiouy"); // vowels, including y
tr->char_plus_apostrophe = empty_wstring;
tr->punct_within_word = punct_in_word;
tr->chars_ignore = chars_ignore_default;
for (ix = 0; ix < 8; ix++) {
tr->stress_amps[ix] = stress_amps2[ix];
tr->stress_lengths[ix] = stress_lengths2[ix];
}
memset(&(tr->langopts), 0, sizeof(tr->langopts));
tr->langopts.max_lengthmod = 500;
tr->langopts.lengthen_tonic = 20;
tr->langopts.stress_rule = STRESSPOSN_2R;
tr->langopts.unstressed_wd1 = 1;
tr->langopts.unstressed_wd2 = 3;
tr->langopts.param[LOPT_SONORANT_MIN] = 95;
tr->langopts.param[LOPT_LONG_VOWEL_THRESHOLD] = 190/2;
tr->langopts.param[LOPT_MAXAMP_EOC] = 19;
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 's'; // don't count this character at start of word
tr->langopts.param[LOPT_BRACKET_PAUSE] = 4; // pause at bracket
tr->langopts.param[LOPT_BRACKET_PAUSE_ANNOUNCED] = 2; // pauses when announcing bracket names
tr->langopts.spelling_stress = false;
tr->langopts.max_initial_consonants = 3;
tr->langopts.replace_chars = NULL;
tr->langopts.alt_alphabet_lang = L('e', 'n');
tr->langopts.roman_suffix = utf8_null;
tr->langopts.lowercase_sentence = false;
SetLengthMods(tr, 201);
tr->langopts.long_stop = 100;
tr->langopts.max_roman = 49;
tr->langopts.min_roman = 2;
tr->langopts.thousands_sep = ',';
tr->langopts.decimal_sep = '.';
tr->langopts.numbers = NUM_DEFAULT;
tr->langopts.break_numbers = BREAK_THOUSANDS;
tr->langopts.max_digits = 14;
// index by 0=. 1=, 2=?, 3=! 4=none, 5=emphasized
unsigned char punctuation_to_tone[INTONATION_TYPES][PUNCT_INTONATIONS] = {
{ 0, 1, 2, 3, 0, 4 },
{ 0, 1, 2, 3, 0, 4 },
{ 5, 6, 2, 3, 0, 4 },
{ 5, 7, 1, 3, 0, 4 },
{ 8, 9, 10, 3, 0, 0 },
{ 8, 8, 10, 3, 0, 0 },
{ 11, 11, 11, 11, 0, 0 }, // 6 test
{ 12, 12, 12, 12, 0, 0 }
};
memcpy(tr->punct_to_tone, punctuation_to_tone, sizeof(tr->punct_to_tone));
memcpy(tr->langopts.tunes, default_tunes, sizeof(tr->langopts.tunes));
return tr;
}
// common letter pairs, encode these as a single byte
// 2 bytes, using the transposed character codes
static const short pairs_ru[] = {
0x010c, // ла 21052 0x23
0x010e, // на 18400
0x0113, // та 14254
0x0301, // ав 31083
0x030f, // ов 13420
0x060e, // не 21798
0x0611, // ре 19458
0x0903, // ви 16226
0x0b01, // ак 14456
0x0b0f, // ок 17836
0x0c01, // ал 13324
0x0c09, // ил 16877
0x0e01, // ан 15359
0x0e06, // ен 13543 0x30
0x0e09, // ин 17168
0x0e0e, // нн 15973
0x0e0f, // он 22373
0x0e1c, // ын 15052
0x0f03, // во 24947
0x0f11, // ро 13552
0x0f12, // со 16368
0x100f, // оп 19054
0x1011, // рп 17067
0x1101, // ар 23967
0x1106, // ер 18795
0x1109, // ир 13797
0x110f, // ор 21737
0x1213, // тс 25076
0x1220, // яс 14310
0x7fff
};
static const unsigned char ru_vowels[] = { // (also kazakh) offset by 0x420 -- а е ё и о у ы э ю я ә ө ұ ү і
0x10, 0x15, 0x31, 0x18, 0x1e, 0x23, 0x2b, 0x2d, 0x2e, 0x2f, 0xb9, 0xc9, 0x91, 0x8f, 0x36, 0
};
static const unsigned char ru_consonants[] = { // б в г д ж з й к л м н п р с т ф х ц ч ш щ ъ ь қ ң һ
0x11, 0x12, 0x13, 0x14, 0x16, 0x17, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1f, 0x20, 0x21, 0x22, 0x24, 0x25, 0x26, 0x27, 0x28, 0x29, 0x2a, 0x2c, 0x73, 0x7b, 0x83, 0x9b, 0
};
static void SetArabicLetters(Translator *tr)
{
const char *arab_vowel_letters = "َ ُ ِ";
const char *arab_consonant_vowel_letters = "ا و ي";
const char *arab_consonant_letters = "ب پ ت ة ث ج ح خ د ذ ر ز س ش ص ض ط ظ ع غ ف ق ك ل م ن ئ ؤ ء أ آ إ ه";
const char *arab_thick_letters = "ص ض ط ظ";
const char *arab_shadda_letter = " ّ ";
const char *arab_hamza_letter = " ّ ";
const char *arab_sukun_letter = " ّ ";
SetLetterBitsUTF8(tr, LETTERGP_A, arab_vowel_letters, OFFSET_ARABIC);
SetLetterBitsUTF8(tr, LETTERGP_B, arab_consonant_vowel_letters, OFFSET_ARABIC);
SetLetterBitsUTF8(tr, LETTERGP_C, arab_consonant_letters, OFFSET_ARABIC);
SetLetterBitsUTF8(tr, LETTERGP_F, arab_thick_letters, OFFSET_ARABIC);
SetLetterBitsUTF8(tr, LETTERGP_G, arab_shadda_letter, OFFSET_ARABIC);
SetLetterBitsUTF8(tr, LETTERGP_H, arab_hamza_letter, OFFSET_ARABIC);
SetLetterBitsUTF8(tr, LETTERGP_Y, arab_sukun_letter, OFFSET_ARABIC);
}
static void SetCyrillicLetters(Translator *tr)
{
// Set letter types for Cyrillic script languages: bg (Bulgarian), ru (Russian), tt (Tatar), uk (Ukrainian).
// character codes offset by 0x420
static const char cyrl_soft[] = { 0x2c, 0x19, 0x27, 0x29, 0 }; // letter group B [k ts; s;] -- ь й ч щ
static const char cyrl_hard[] = { 0x2a, 0x16, 0x26, 0x28, 0 }; // letter group H [S Z ts] -- ъ ж ц ш
static const char cyrl_nothard[] = { 0x11, 0x12, 0x13, 0x14, 0x17, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1f, 0x20, 0x21, 0x22, 0x24, 0x25, 0x27, 0x29, 0x2c, 0 }; // б в г д з й к л м н п р с т ф х ч щ ь
static const char cyrl_voiced[] = { 0x11, 0x12, 0x13, 0x14, 0x16, 0x17, 0 }; // letter group G (voiced obstruents) -- б в г д ж з
static const char cyrl_ivowels[] = { 0x2c, 0x2e, 0x2f, 0x31, 0 }; // letter group Y (iotated vowels & soft-sign) -- ь ю я ё
tr->encoding = ESPEAKNG_ENCODING_KOI8_R;
tr->transpose_min = 0x430; // convert cyrillic from unicode into range 0x01 to 0x22
tr->transpose_max = 0x451;
tr->transpose_map = NULL;
tr->frequent_pairs = pairs_ru;
tr->letter_bits_offset = OFFSET_CYRILLIC;
memset(tr->letter_bits, 0, sizeof(tr->letter_bits));
SetLetterBits(tr, LETTERGP_A, (char *)ru_vowels);
SetLetterBits(tr, LETTERGP_B, cyrl_soft);
SetLetterBits(tr, LETTERGP_C, (char *)ru_consonants);
SetLetterBits(tr, LETTERGP_H, cyrl_hard);
SetLetterBits(tr, LETTERGP_F, cyrl_nothard);
SetLetterBits(tr, LETTERGP_G, cyrl_voiced);
SetLetterBits(tr, LETTERGP_Y, cyrl_ivowels);
SetLetterBits(tr, LETTERGP_VOWEL2, (char *)ru_vowels);
}
static void SetIndicLetters(Translator *tr)
{
// Set letter types for Devanagari (Indic) script languages: Devanagari, Tamill, etc.
static const char deva_consonants2[] = { 0x02, 0x03, 0x58, 0x59, 0x5a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f, 0x7b, 0x7c, 0x7e, 0x7f, 0 };
static const char deva_vowels2[] = { 0x60, 0x61, 0x55, 0x56, 0x57, 0x62, 0x63, 0 }; // non-consecutive vowels and vowel-signs
memset(tr->letter_bits, 0, sizeof(tr->letter_bits));
SetLetterBitsRange(tr, LETTERGP_A, 0x04, 0x14); // vowel letters
SetLetterBitsRange(tr, LETTERGP_A, 0x3e, 0x4d); // + vowel signs, and virama
SetLetterBits(tr, LETTERGP_A, deva_vowels2); // + extra vowels and vowel signs
SetLetterBitsRange(tr, LETTERGP_B, 0x3e, 0x4d); // vowel signs, and virama
SetLetterBits(tr, LETTERGP_B, deva_vowels2); // + extra vowels and vowel signs
SetLetterBitsRange(tr, LETTERGP_C, 0x15, 0x39); // the main consonant range
SetLetterBits(tr, LETTERGP_C, deva_consonants2); // + additional consonants
SetLetterBitsRange(tr, LETTERGP_Y, 0x04, 0x14); // vowel letters
SetLetterBitsRange(tr, LETTERGP_Y, 0x3e, 0x4c); // + vowel signs
SetLetterBits(tr, LETTERGP_Y, deva_vowels2); // + extra vowels and vowel signs
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 1; // disable check for unpronouncable words
tr->langopts.suffix_add_e = tr->letter_bits_offset + 0x4d; // virama
}
static void SetupTranslator(Translator *tr, const short *lengths, const unsigned char *amps)
{
if (lengths != NULL)
memcpy(tr->stress_lengths, lengths, sizeof(tr->stress_lengths));
if (amps != NULL)
memcpy(tr->stress_amps, amps, sizeof(tr->stress_amps));
}
Translator *SelectTranslator(const char *name)
{
int name2 = 0;
Translator *tr;
static const short stress_lengths_equal[8] = { 230, 230, 230, 230, 0, 0, 230, 230 };
static const unsigned char stress_amps_equal[8] = { 19, 19, 19, 19, 19, 19, 19, 19 };
static const short stress_lengths_fr[8] = { 190, 170, 190, 200, 0, 0, 190, 240 };
static const unsigned char stress_amps_fr[8] = { 18, 16, 18, 18, 18, 18, 18, 18 };
static const unsigned char stress_amps_sk[8] = { 17, 16, 20, 20, 20, 22, 22, 21 };
static const short stress_lengths_sk[8] = { 190, 190, 210, 210, 0, 0, 210, 210 };
static const short stress_lengths_ta[8] = { 200, 200, 210, 210, 0, 0, 230, 230 };
static const short stress_lengths_ta2[8] = { 230, 230, 240, 240, 0, 0, 260, 260 };
static const unsigned char stress_amps_ta[8] = { 18, 18, 18, 18, 20, 20, 22, 22 };
tr = NewTranslator();
strcpy(tr->dictionary_name, name);
// convert name string into a word of up to 4 characters, for the switch()
while (*name != 0)
name2 = (name2 << 8) + *name++;
switch (name2)
{
case L('m', 'i'):
case L('m', 'y'):
case L4('p', 'i', 'q', 'd'): // piqd
case L('p', 'y'):
case L('q', 'u'):
case L3('q', 'u', 'c'):
case L('t', 'h'):
case L('u', 'z'):
{
tr->langopts.numbers = 0; // disable numbers until the definition are complete in _list file
}
break;
case L('a', 'f'):
{
static const short stress_lengths_af[8] = { 170, 140, 220, 220, 0, 0, 250, 270 };
SetupTranslator(tr, stress_lengths_af, NULL);
tr->langopts.stress_rule = STRESSPOSN_1L;
tr->langopts.vowel_pause = 0x30;
tr->langopts.param[LOPT_DIERESES] = 1;
tr->langopts.param[LOPT_PREFIXES] = 1;
SetLetterVowel(tr, 'y'); // add 'y' to vowels
tr->langopts.numbers = NUM_SWAP_TENS | NUM_HUNDRED_AND | NUM_SINGLE_AND | NUM_ROMAN | NUM_1900;
tr->langopts.accents = 1;
}
break;
case L('a', 'm'): // Amharic, Ethiopia
{
SetupTranslator(tr, stress_lengths_fr, stress_amps_fr);
tr->letter_bits_offset = OFFSET_ETHIOPIC;
tr->langopts.stress_rule = STRESSPOSN_1L;
tr->langopts.stress_flags = S_NO_AUTO_2 | S_FINAL_DIM; // don't use secondary stress
tr->langopts.length_mods0 = tr->langopts.length_mods; // don't lengthen vowels in the last syllable
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 1; // disable check for unpronouncable words
tr->langopts.numbers = NUM_OMIT_1_HUNDRED;
}
break;
case L('a', 'r'): // Arabic
tr->transpose_min = OFFSET_ARABIC; // for ar_list, use 6-bit character codes
tr->transpose_max = 0x65f;
tr->transpose_map = NULL;
tr->letter_bits_offset = OFFSET_ARABIC;
tr->langopts.numbers = NUM_SWAP_TENS | NUM_AND_UNITS | NUM_HUNDRED_AND | NUM_OMIT_1_HUNDRED | NUM_AND_HUNDRED | NUM_THOUSAND_AND | NUM_OMIT_1_THOUSAND;
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 1; // disable check for unpronouncable words
tr->encoding = ESPEAKNG_ENCODING_ISO_8859_6;
SetArabicLetters(tr);
break;
case L('b', 'e'): // Belarusian
{
static const unsigned char stress_amps_be[8] = { 12, 10, 8, 8, 0, 0, 16, 17 };
static const short stress_lengths_be[8] = { 160, 140, 200, 140, 0, 0, 240, 160 };
static const wchar_t vowels_be[] = { // offset by 0x420 -- а е ё о у ы э ю я і
0x10, 0x15, 0x31, 0x1e, 0x23, 0x2b, 0x2d, 0x2e, 0x2f, 0x36, 0
};
static const unsigned char consonants_be[] = { // б в г д ж з й к л м н п р с т ф х ц ч ш ў
0x11, 0x12, 0x13, 0x14, 0x16, 0x17, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1f, 0x20, 0x21, 0x22, 0x24, 0x25, 0x26, 0x27, 0x28, 0x3e, 0
};
tr->langopts.stress_flags = S_NO_AUTO_2 | S_NO_DIM; // don't use secondary stress
tr->letter_bits_offset = OFFSET_CYRILLIC;
tr->transpose_min = 0x430; // convert cyrillic from unicode into range 0x01 to 0x2f
tr->transpose_max = 0x45e;
memset(tr->letter_bits, 0, sizeof(tr->letter_bits));
SetLetterBits(tr, LETTERGP_A, (char *)vowels_be);
SetLetterBits(tr, LETTERGP_C, (char *)consonants_be);
SetLetterBits(tr, LETTERGP_VOWEL2, (char *)vowels_be);
SetupTranslator(tr, stress_lengths_be, stress_amps_be);
tr->encoding = ESPEAKNG_ENCODING_ISO_8859_5;
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 1; // disable check for unpronouncable words
tr->langopts.stress_rule = STRESSPOSN_1L;
tr->langopts.numbers = NUM_DECIMAL_COMMA | NUM_OMIT_1_HUNDRED;
tr->langopts.numbers2 = NUM2_THOUSANDPLEX_VAR_THOUSANDS | NUM2_THOUSANDS_VAR1; // variant numbers before thousands
}
break;
case L('b', 'g'): // Bulgarian
{
SetCyrillicLetters(tr);
SetLetterVowel(tr, 0x2a);
tr->encoding = ESPEAKNG_ENCODING_ISO_8859_5;
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 0x432; // [v] don't count this character at start of word
tr->langopts.param[LOPT_REGRESSIVE_VOICING] = 0x107; // devoice at end of word, and change voicing to match a following consonant (except v)
tr->langopts.param[LOPT_REDUCE] = 2;
tr->langopts.stress_rule = STRESSPOSN_2R;
tr->langopts.numbers = NUM_DECIMAL_COMMA | NUM_ALLOW_SPACE | NUM_OMIT_1_HUNDRED | NUM_HUNDRED_AND | NUM_AND_UNITS | NUM_SINGLE_AND | NUM_ROMAN | NUM_ROMAN_ORDINAL | NUM_ROMAN_CAPITALS;
tr->langopts.thousands_sep = ' '; // don't allow dot as thousands separator
}
break;
case L('b', 'n'): // Bengali
case L('a', 's'): // Assamese
case L3('b', 'p', 'y'): // Manipuri (temporary placement - it's not indo-european)
{
static const short stress_lengths_bn[8] = { 180, 180, 210, 210, 0, 0, 230, 240 };
static const unsigned char stress_amps_bn[8] = { 18, 18, 18, 18, 20, 20, 22, 22 };
static const char bn_consonants2[3] = { 0x70, 0x71, 0 };
SetupTranslator(tr, stress_lengths_bn, stress_amps_bn);
tr->langopts.length_mods0 = tr->langopts.length_mods; // don't lengthen vowels in the last syllable
tr->langopts.stress_rule = STRESSPOSN_1L;
tr->langopts.stress_flags = S_MID_DIM | S_FINAL_DIM; // use 'diminished' for unstressed final syllable
tr->letter_bits_offset = OFFSET_BENGALI;
SetIndicLetters(tr); // call this after setting OFFSET_BENGALI
SetLetterBitsRange(tr, LETTERGP_B, 0x01, 0x01); // candranindu
SetLetterBitsRange(tr, LETTERGP_F, 0x3e, 0x4c); // vowel signs, but not virama
SetLetterBits(tr, LETTERGP_C, bn_consonants2);
tr->langopts.numbers = NUM_SWAP_TENS;
tr->langopts.break_numbers = BREAK_LAKH_BN;
if (name2 == L3('b', 'p', 'y')) {
tr->langopts.numbers = NUM_DEFAULT;
tr->langopts.numbers2 = NUM2_SWAP_THOUSANDS;
}
}
break;
case L3('c', 'h', 'r'): // Cherokee
{
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 1; // disable check for unpronouncable words
tr->langopts.stress_flags = S_NO_AUTO_2 | S_FINAL_DIM | S_FINAL_DIM_ONLY | S_EO_CLAUSE1;
}
break;
case L('c', 'y'): // Welsh
{
static const short stress_lengths_cy[8] = { 170, 220, 180, 180, 0, 0, 250, 270 };
static const unsigned char stress_amps_cy[8] = { 17, 15, 18, 18, 0, 0, 22, 20 }; // 'diminished' is used to mark a quieter, final unstressed syllable
SetupTranslator(tr, stress_lengths_cy, stress_amps_cy);
tr->encoding = ESPEAKNG_ENCODING_ISO_8859_14;
tr->langopts.stress_rule = STRESSPOSN_2R;
// 'diminished' is an unstressed final syllable
tr->langopts.stress_flags = S_FINAL_DIM_ONLY | S_FINAL_NO_2;
tr->langopts.unstressed_wd1 = 0;
tr->langopts.unstressed_wd2 = 2;
tr->langopts.param[LOPT_SONORANT_MIN] = 120; // limit the shortening of sonorants before short vowels
tr->langopts.numbers = NUM_OMIT_1_HUNDRED;
SetLetterVowel(tr, 'w'); // add letter to vowels and remove from consonants
SetLetterVowel(tr, 'y');
}
break;
case L('d', 'a'): // Danish
{
static const short stress_lengths_da[8] = { 160, 140, 200, 200, 0, 0, 220, 230 };
SetupTranslator(tr, stress_lengths_da, NULL);
tr->langopts.stress_rule = STRESSPOSN_1L;
tr->langopts.param[LOPT_PREFIXES] = 1;
SetLetterVowel(tr, 'y');
tr->langopts.numbers = NUM_DECIMAL_COMMA | NUM_SWAP_TENS | NUM_HUNDRED_AND | NUM_OMIT_1_HUNDRED | NUM_ORDINAL_DOT | NUM_1900 | NUM_ROMAN | NUM_ROMAN_CAPITALS | NUM_ROMAN_ORDINAL;
}
break;
case L('d', 'e'):
{
static const short stress_lengths_de[8] = { 150, 130, 200, 200, 0, 0, 270, 270 };
static const unsigned char stress_amps_de[] = { 20, 20, 20, 20, 20, 22, 22, 20 };
SetupTranslator(tr, stress_lengths_de, stress_amps_de);
tr->langopts.stress_rule = STRESSPOSN_1L;
tr->langopts.word_gap = 0x8; // don't use linking phonemes
tr->langopts.vowel_pause = 0x30;
tr->langopts.param[LOPT_PREFIXES] = 1;
tr->langopts.param[LOPT_REGRESSIVE_VOICING] = 0x100; // devoice at end of word
tr->langopts.param[LOPT_LONG_VOWEL_THRESHOLD] = 175/2;
tr->langopts.numbers = NUM_DECIMAL_COMMA | NUM_SWAP_TENS | NUM_ALLOW_SPACE | NUM_ORDINAL_DOT | NUM_ROMAN;
SetLetterVowel(tr, 'y');
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 2; // use de_rules for unpronouncable rules
}
break;
case L('e', 'n'):
{
static const short stress_lengths_en[8] = { 182, 140, 220, 220, 0, 0, 248, 275 };
SetupTranslator(tr, stress_lengths_en, NULL);
tr->langopts.stress_rule = STRESSPOSN_1L;
tr->langopts.stress_flags = 0x08;
tr->langopts.numbers = NUM_HUNDRED_AND | NUM_ROMAN | NUM_1900;
tr->langopts.max_digits = 33;
tr->langopts.param[LOPT_COMBINE_WORDS] = 2; // allow "mc" to cmbine with the following word
tr->langopts.suffix_add_e = 'e';
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 2; // use en_rules for unpronouncable rules
SetLetterBits(tr, 6, "aeiouy"); // Group Y: vowels, including y
}
break;
case L('e', 'l'): // Greek
case L3('g', 'r', 'c'): // Ancient Greek
{
static const short stress_lengths_el[8] = { 155, 180, 210, 210, 0, 0, 270, 300 };
static const unsigned char stress_amps_el[8] = { 15, 12, 20, 20, 20, 22, 22, 21 }; // 'diminished' is used to mark a quieter, final unstressed syllable
// character codes offset by 0x380
static const char el_vowels[] = { 0x10, 0x2c, 0x2d, 0x2e, 0x2f, 0x30, 0x31, 0x35, 0x37, 0x39, 0x3f, 0x45, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, 0 };
static const char el_fvowels[] = { 0x2d, 0x2e, 0x2f, 0x35, 0x37, 0x39, 0x45, 0x4d, 0 }; // ε η ι υ έ ή ί ύ _
static const char el_voiceless[] = { 0x38, 0x3a, 0x3e, 0x40, 0x42, 0x43, 0x44, 0x46, 0x47, 0 }; // θ κ ξ π ς σ τ φ χ _
static const char el_consonants[] = { 0x32, 0x33, 0x34, 0x36, 0x38, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x40, 0x41, 0x42, 0x43, 0x44, 0x46, 0x47, 0x48, 0 };
static const wchar_t el_char_apostrophe[] = { 0x3c3, 0 }; // σ _
SetupTranslator(tr, stress_lengths_el, stress_amps_el);
tr->encoding = ESPEAKNG_ENCODING_ISO_8859_7;
tr->char_plus_apostrophe = el_char_apostrophe;
tr->letter_bits_offset = OFFSET_GREEK;
memset(tr->letter_bits, 0, sizeof(tr->letter_bits));
SetLetterBits(tr, LETTERGP_A, el_vowels);
SetLetterBits(tr, LETTERGP_VOWEL2, el_vowels);
SetLetterBits(tr, LETTERGP_B, el_voiceless);
SetLetterBits(tr, LETTERGP_C, el_consonants);
SetLetterBits(tr, LETTERGP_Y, el_fvowels); // front vowels: ε η ι υ _
tr->langopts.length_mods0 = tr->langopts.length_mods; // don't lengthen vowels in the last syllable
tr->langopts.stress_rule = STRESSPOSN_2R;
tr->langopts.stress_flags = S_FINAL_DIM_ONLY; // mark unstressed final syllables as diminished
tr->langopts.unstressed_wd1 = 0;
tr->langopts.unstressed_wd2 = 2;
tr->langopts.param[LOPT_SONORANT_MIN] = 130; // limit the shortening of sonorants before short vowels
tr->langopts.numbers = NUM_SINGLE_STRESS | NUM_DECIMAL_COMMA;
tr->langopts.numbers2 = NUM2_THOUSANDPLEX_VAR_THOUSANDS | NUM2_MULTIPLE_ORDINAL | NUM2_ORDINAL_NO_AND;
if (name2 == L3('g', 'r', 'c')) {
// ancient greek
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 1;
}
}
break;
case L('e', 'o'):
{
static const short stress_lengths_eo[8] = { 150, 140, 180, 180, 0, 0, 200, 200 };
static const unsigned char stress_amps_eo[] = { 16, 14, 20, 20, 20, 22, 22, 21 };
static const wchar_t eo_char_apostrophe[2] = { 'l', 0 };
SetupTranslator(tr, stress_lengths_eo, stress_amps_eo);
tr->encoding = ESPEAKNG_ENCODING_ISO_8859_3;
tr->char_plus_apostrophe = eo_char_apostrophe;
tr->langopts.vowel_pause = 2;
tr->langopts.stress_rule = STRESSPOSN_2R;
tr->langopts.stress_flags = S_FINAL_DIM_ONLY | S_FINAL_NO_2;
tr->langopts.unstressed_wd2 = 2;
tr->langopts.numbers = NUM_DECIMAL_COMMA | NUM_OMIT_1_HUNDRED | NUM_ALLOW_SPACE | NUM_ROMAN;
}
break;
case L('e', 's'): // Spanish
case L('a', 'n'): // Aragonese
case L('c', 'a'): // Catalan
case L('i', 'a'): // Interlingua
case L3('p', 'a', 'p'): // Papiamento
{
static const short stress_lengths_es[8] = { 160, 145, 155, 150, 0, 0, 200, 245 };
static const unsigned char stress_amps_es[8] = { 16, 14, 15, 16, 20, 20, 22, 22 }; // '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
SetupTranslator(tr, stress_lengths_es, stress_amps_es);
tr->langopts.length_mods0 = tr->langopts.length_mods; // don't lengthen vowels in the last syllable
tr->langopts.stress_rule = STRESSPOSN_2R;
// stress last syllable if it doesn't end in vowel or "s" or "n"
// 'diminished' is an unstressed final syllable
tr->langopts.stress_flags = S_FINAL_SPANISH | S_FINAL_DIM_ONLY | S_FINAL_NO_2;
tr->langopts.unstressed_wd1 = 0;
tr->langopts.unstressed_wd2 = 2;
tr->langopts.param[LOPT_SONORANT_MIN] = 120; // limit the shortening of sonorants before short vowels
tr->langopts.numbers = NUM_SINGLE_STRESS | NUM_DECIMAL_COMMA | NUM_AND_UNITS | NUM_OMIT_1_HUNDRED | NUM_OMIT_1_THOUSAND | NUM_ROMAN | NUM_ROMAN_AFTER | NUM_DFRACTION_4;
tr->langopts.numbers2 = NUM2_MULTIPLE_ORDINAL | NUM2_ORDINAL_NO_AND;
if (name2 == L('c', 'a')) {
// stress last syllable unless word ends with a vowel
tr->punct_within_word = ca_punct_within_word;
tr->langopts.stress_flags = S_FINAL_SPANISH | S_FINAL_DIM_ONLY | S_FINAL_NO_2 | S_NO_AUTO_2 | S_FIRST_PRIMARY;
} else if (name2 == L('i', 'a')) {
tr->langopts.stress_flags = S_FINAL_SPANISH | S_FINAL_DIM_ONLY | S_FINAL_NO_2;
tr->langopts.numbers = NUM_DECIMAL_COMMA | NUM_OMIT_1_HUNDRED | NUM_OMIT_1_THOUSAND | NUM_ROMAN | NUM_ROMAN_AFTER;
} else if (name2 == L('a', 'n')) {
tr->langopts.stress_flags = S_FINAL_SPANISH | S_FINAL_DIM_ONLY | S_FINAL_NO_2;
tr->langopts.numbers = NUM_SINGLE_STRESS | NUM_DECIMAL_COMMA | NUM_AND_UNITS | NUM_OMIT_1_HUNDRED | NUM_OMIT_1_THOUSAND | NUM_ROMAN | NUM_ROMAN_ORDINAL;
tr->langopts.numbers2 = NUM2_ORDINAL_NO_AND;
tr->langopts.roman_suffix = utf8_ordinal;
} else if (name2 == L3('p', 'a', 'p')) {
// stress last syllable unless word ends with a vowel
tr->langopts.stress_rule = STRESSPOSN_1R;
tr->langopts.stress_flags = S_FINAL_VOWEL_UNSTRESSED | S_FINAL_DIM_ONLY | S_FINAL_NO_2 | S_NO_AUTO_2;
} else
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 2; // use es_rules for unpronouncable rules
}
break;
case L('e', 'u'): // basque
{
static const short stress_lengths_eu[8] = { 200, 200, 200, 200, 0, 0, 210, 230 }; // very weak stress
static const unsigned char stress_amps_eu[8] = { 16, 16, 18, 18, 18, 18, 18, 18 };
SetupTranslator(tr, stress_lengths_eu, stress_amps_eu);
tr->langopts.stress_flags = S_FINAL_VOWEL_UNSTRESSED | S_MID_DIM;
tr->langopts.param[LOPT_SUFFIX] = 1;
tr->langopts.numbers = NUM_SINGLE_STRESS | NUM_DECIMAL_COMMA | NUM_HUNDRED_AND | NUM_OMIT_1_HUNDRED | NUM_OMIT_1_THOUSAND | NUM_VIGESIMAL;
}
break;
case L('f', 'a'): // Farsi
{
// Convert characters in the range 0x620 to 0x6cc to the range 1 to 63.
// 0 indicates no translation for this character
static const char transpose_map_fa[] = {
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, // 0x620
16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 0, 0, 0, 0, 0, // 0x630
0, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, // 0x640
42, 43, 0, 0, 44, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x650
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x660
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 45, 0, // 0x670
0, 0, 0, 0, 0, 0, 46, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x680
0, 0, 0, 0, 0, 0, 0, 0, 47, 0, 0, 0, 0, 0, 0, 0, // 0x690
0, 0, 0, 0, 0, 0, 0, 0, 0, 48, 0, 0, 0, 0, 0, 49, // 0x6a0
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 0x6b0
50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 51 // 0x6c0
};
tr->transpose_min = 0x620;
tr->transpose_max = 0x6cc;
tr->transpose_map = transpose_map_fa;
tr->letter_bits_offset = OFFSET_ARABIC;
tr->langopts.numbers = NUM_AND_UNITS | NUM_HUNDRED_AND;
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 1; // disable check for unpronouncable words
tr->chars_ignore = chars_ignore_zwnj_hyphen; // replace ZWNJ by hyphen
}
break;
case L('e', 't'): // Estonian
tr->encoding = ESPEAKNG_ENCODING_ISO_8859_4;
// fallthrough:
case L('f', 'i'): // Finnish
{
tr->langopts.long_stop = 130;
tr->langopts.numbers = NUM_DECIMAL_COMMA | NUM_ALLOW_SPACE | NUM_DFRACTION_2 | NUM_ORDINAL_DOT;
SetLetterVowel(tr, 'y');
}
break;
case L('f', 'o'): // Faroese
{
//static const short stress_lengths_da[8] = { 160, 140, 200, 200, 0, 0, 220, 230 };
//SetupTranslator(tr, stress_lengths_da, NULL);
//tr->langopts.stress_rule = STRESSPOSN_1L;
//tr->langopts.param[LOPT_PREFIXES] = 1;
//SetLetterVowel(tr, 'y');
tr->langopts.numbers = NUM_DECIMAL_COMMA | NUM_SWAP_TENS | NUM_HUNDRED_AND | NUM_OMIT_1_HUNDRED | NUM_ORDINAL_DOT | NUM_1900 | NUM_ROMAN | NUM_ROMAN_CAPITALS | NUM_ROMAN_ORDINAL;
}
break;
case L('f', 'r'): // french
{
SetupTranslator(tr, stress_lengths_fr, stress_amps_fr);
tr->langopts.stress_rule = STRESSPOSN_1R; // stress on final syllable
tr->langopts.stress_flags = S_NO_AUTO_2 | S_FINAL_DIM; // don't use secondary stress
tr->langopts.param[LOPT_IT_LENGTHEN] = 1; // remove lengthen indicator from unstressed syllables
tr->langopts.length_mods0 = tr->langopts.length_mods; // don't lengthen vowels in the last syllable
tr->langopts.accents = 2; // Say "Capital" after the letter.
tr->langopts.numbers = NUM_SINGLE_STRESS | NUM_DECIMAL_COMMA | NUM_ALLOW_SPACE | NUM_OMIT_1_HUNDRED | NUM_NOPAUSE | NUM_ROMAN | NUM_ROMAN_CAPITALS | NUM_ROMAN_AFTER | NUM_VIGESIMAL | NUM_DFRACTION_4;
SetLetterVowel(tr, 'y');
}
break;
case L3('h','a', 'k'): // Hakka Chinese
{
tr->langopts.stress_flags = S_NO_DIM; // don't automatically set diminished stress (may be set in the intonation module)
tr->langopts.tone_language = 1; // Tone language, use CalcPitches_Tone() rather than CalcPitches()
tr->langopts.tone_numbers = 1; // a number after letters indicates a tone number (eg. pinyin or jyutping)
tr->langopts.ideographs = 1;
}
break;
case L('g', 'a'): // irish
case L('g', 'd'): // scots gaelic
{
tr->langopts.stress_rule = STRESSPOSN_1L;
tr->langopts.stress_flags = S_NO_AUTO_2; // don't use secondary stress
tr->langopts.numbers = NUM_OMIT_1_HUNDRED | NUM_OMIT_1_THOUSAND;
tr->langopts.accents = 2; // 'capital' after letter name
tr->langopts.param[LOPT_UNPRONOUNCABLE] = 3; // don't count apostrophe
tr->langopts.param[LOPT_IT_LENGTHEN] = 1; // remove [:] phoneme from non-stressed syllables (Lang=gd)
}
break;
case L('g','n'): // guarani
{
tr->langopts.stress_rule = STRESSPOSN_1R; // stress on final syllable
tr->langopts.length_mods0 = tr->langopts.length_mods; // don't lengthen vowels in the last syllable
}
break;
case L('h', 'i'): // Hindi
case L('n', 'e'): // Nepali
case L('o', 'r'): // Oriya
case L('p', 'a'): // Punjabi
case L('g', 'u'): // Gujarati
case L('m', 'r'): // Marathi
{
static const short stress_lengths_hi[8] = { 190, 190, 210, 210, 0, 0, 230, 250 };
static const unsigned char stress_amps_hi[8] = { 17, 14, 20, 19, 20, 22, 22, 21 };
SetupTranslator(tr, stress_lengths_hi, stress_amps_hi);
tr->encoding = ESPEAKNG_ENCODING_ISCII;
tr->langopts.length_mods0 = tr->langopts.length_mods; // don't lengthen vowels in the last syllable
tr->langopts.stress_rule = STRESSPOSN_1RH; // stress on last heaviest syllable, excluding final syllable
tr->langopts.stress_flags = S_MID_DIM | S_FINAL_DIM; // use 'diminished' for unstressed final syllable
tr->langopts.numbers = NUM_SWAP_TENS;
tr->langopts.break_numbers = BREAK_LAKH_HI;
tr->letter_bits_offset = OFFSET_DEVANAGARI;
if (name2 == L('p', 'a'))
tr->letter_bits_offset = OFFSET_GURMUKHI;
else if (name2 == L('g', 'u')) {
SetupTranslator(tr, stress_lengths_equal, stress_amps_equal);
tr->letter_bits_offset = OFFSET_GUJARATI;
tr->langopts.stress_rule = STRESSPOSN_2R;
} else if (name2 == L('n', 'e')) {
SetupTranslator(tr, stress_lengths_equal, stress_amps_equal);
tr->langopts.break_numbers = BREAK_LAKH;
tr->langopts.max_digits = 22;
tr->langopts.numbers2 |= NUM2_ENGLISH_NUMERALS;
} else if (name2 == L('o', 'r'))
tr->letter_bits_offset = OFFSET_ORIYA;
SetIndicLetters(tr);
}
break;
case L('h', 'r'): // Croatian
case L('b', 's'): // Bosnian
case L('s', 'r'): // Serbian
{
static const unsigned char stress_amps_hr[8] = { 17, 17, 20, 20, 20, 22, 22, 21 };
static const short stress_lengths_hr[8] = { 180, 160, 200, 200, 0, 0, 220, 230 };
static const short stress_lengths_sr[8] = { 160, 150, 200, 200, 0, 0, 250, 260 };
strcpy(tr->dictionary_name, "hbs");
if (name2 == L('s', 'r'))
SetupTranslator(tr, stress_lengths_sr, stress_amps_hr);
else
SetupTranslator(tr, stress_lengths_hr, stress_amps_hr);
tr->encoding = ESPEAKNG_ENCODING_ISO_8859_2;
tr->langopts.stress_rule = STRESSPOSN_1L;
tr->langopts.stress_flags = S_FINAL_NO_2;
tr->langopts.param[LOPT_REGRESSIVE_VOICING] = 0x3;
tr->langopts.max_initial_consonants = 5;
tr->langopts.spelling_stress = true;
tr->langopts.accents = 1;
tr->langopts.numbers = NUM_SINGLE_STRESS | NUM_HUNDRED_AND | NUM_OMIT_1_HUNDRED | NUM_DECIMAL_COMMA | NUM_THOUS_SPACE | NUM_DFRACTION_2 | NUM_ROMAN_CAPITALS;
tr->langopts.numbers2 = NUM2_THOUSANDPLEX_VAR_THOUSANDS | NUM2_THOUSANDPLEX_VAR_MILLIARDS | NUM2_THOUSANDS_VAR5;
tr->langopts.our_alphabet = OFFSET_CYRILLIC; // don't say "cyrillic" before letter names
SetLetterVowel(tr, 'y');
SetLetterVowel(tr, 'r');
}
break;
case L('h', 't'): // Haitian Creole
tr->langopts.stress_rule = STRESSPOSN_1R; // stress on final syllable
tr->langopts.stress_flags = S_NO_AUTO_2 | S_FINAL_DIM; // don't use secondary stress
tr->langopts.numbers = NUM_SINGLE_STRESS | NUM_OMIT_1_HUNDRED | NUM_NOPAUSE | NUM_ROMAN | NUM_VIGESIMAL | NUM_DFRACTION_4;
break;
case L('h', 'u'): // Hungarian
{
static const unsigned char stress_amps_hu[8] = { 17, 17, 19, 19, 20, 22, 22, 21 };
static const short stress_lengths_hu[8] = { 185, 195, 195, 190, 0, 0, 210, 220 };
SetupTranslator(tr, stress_lengths_hu, stress_amps_hu);
tr->encoding = ESPEAKNG_ENCODING_ISO_8859_2;
tr->langopts.vowel_pause = 0x20;
tr->langopts.stress_rule = STRESSPOSN_1L;
tr->langopts.stress_flags = S_FINAL_DIM_ONLY | S_FINAL_NO_2 | S_NO_AUTO_2 | 0x8000 | S_HYPEN_UNSTRESS;
tr->langopts.unstressed_wd1 = 2;
tr->langopts.param[LOPT_ANNOUNCE_PUNCT] = 2; // don't break clause before announcing . ? !
tr->langopts.numbers = NUM_DFRACTION_5 | NUM_ALLOW_SPACE | NUM_ROMAN | NUM_ROMAN_ORDINAL | NUM_ROMAN_CAPITALS | NUM_ORDINAL_DOT | NUM_OMIT_1_HUNDRED | NUM_OMIT_1_THOUSAND;
tr->langopts.thousands_sep = ' '; // don't allow dot as thousands separator
tr->langopts.decimal_sep = ',';
tr->langopts.max_roman = 899;
tr->langopts.min_roman = 1;
SetLetterVowel(tr, 'y');
tr->langopts.spelling_stress = true;
SetLengthMods(tr, 3); // all equal
}
break;
case L('h', 'y'): // Armenian
{
static const short stress_lengths_hy[8] = { 250, 200, 250, 250, 0, 0, 250, 250 };
static const char hy_vowels[] = { 0x31, 0x35, 0x37, 0x38, 0x3b, 0x48, 0x55, 0 };
static const char hy_consonants[] = {
0x32, 0x33, 0x34, 0x36, 0x39, 0x3a, 0x3c, 0x3d, 0x3e, 0x3f, 0x40, 0x41, 0x42, 0x43, 0x44,
0x46, 0x47, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, 0x50, 0x51, 0x52, 0x53, 0x54, 0x56, 0
};
static const char hy_consonants2[] = { 0x45, 0 };
SetupTranslator(tr, stress_lengths_hy, NULL);
tr->langopts.stress_rule = STRESSPOSN_1R; // default stress on final syllable
tr->letter_bits_offset = OFFSET_ARMENIAN;
memset(tr->letter_bits, 0, sizeof(tr->letter_bits));
SetLetterBits(tr, LETTERGP_A, hy_vowels);
SetLetterBits(tr, LETTERGP_VOWEL2, hy_vowels);
SetLetterBits(tr, LETTERGP_B, hy_consonants); // not including 'j'
SetLetterBits(tr, LETTERGP_C, hy_consonants);