eSpeak NG is an open source speech synthesizer that supports more than hundred languages and accents.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

espeak-ng.c 20KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724
  1. /*
  2. * Copyright (C) 2006 to 2013 by Jonathan Duddington
  3. * email: [email protected]
  4. * Copyright (C) 2015-2016 Reece H. Dunn
  5. *
  6. * This program is free software; you can redistribute it and/or modify
  7. * it under the terms of the GNU General Public License as published by
  8. * the Free Software Foundation; either version 3 of the License, or
  9. * (at your option) any later version.
  10. *
  11. * This program is distributed in the hope that it will be useful,
  12. * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13. * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  14. * GNU General Public License for more details.
  15. *
  16. * You should have received a copy of the GNU General Public License
  17. * along with this program; if not, see: <http://www.gnu.org/licenses/>.
  18. */
  19. #include "config.h"
  20. #include <ctype.h>
  21. #include <errno.h>
  22. #include <getopt.h>
  23. #include <stdbool.h>
  24. #include <stdio.h>
  25. #include <stdlib.h>
  26. #include <string.h>
  27. #include <sys/stat.h>
  28. #include <time.h>
  29. #include <espeak-ng/espeak_ng.h>
  30. #include <espeak/speak_lib.h>
  31. #ifndef PROGRAM_NAME
  32. #define PROGRAM_NAME "espeak-ng"
  33. #endif
  34. #ifndef PLAYBACK_MODE
  35. #define PLAYBACK_MODE ENOUTPUT_MODE_SPEAK_AUDIO
  36. #endif
  37. extern ESPEAK_NG_API void strncpy0(char *to, const char *from, int size);
  38. extern ESPEAK_NG_API int utf8_in(int *c, const char *buf);
  39. extern ESPEAK_NG_API int GetFileLength(const char *filename);
  40. static const char *help_text =
  41. "\n" PROGRAM_NAME " [options] [\"<words>\"]\n\n"
  42. "-f <text file> Text file to speak\n"
  43. "--stdin Read text input from stdin instead of a file\n\n"
  44. "If neither -f nor --stdin, then <words> are spoken, or if none then text\n"
  45. "is spoken from stdin, each line separately.\n\n"
  46. "-a <integer>\n"
  47. "\t Amplitude, 0 to 200, default is 100\n"
  48. "-d <device>\n"
  49. "\t Use the specified device to speak the audio on. If not specified, the\n"
  50. "\t default audio device is used.\n"
  51. "-g <integer>\n"
  52. "\t Word gap. Pause between words, units of 10mS at the default speed\n"
  53. "-k <integer>\n"
  54. "\t Indicate capital letters with: 1=sound, 2=the word \"capitals\",\n"
  55. "\t higher values indicate a pitch increase (try -k20).\n"
  56. "-l <integer>\n"
  57. "\t Line length. If not zero (which is the default), consider\n"
  58. "\t lines less than this length as end-of-clause\n"
  59. "-p <integer>\n"
  60. "\t Pitch adjustment, 0 to 99, default is 50\n"
  61. "-s <integer>\n"
  62. "\t Speed in approximate words per minute. The default is 175\n"
  63. "-v <voice name>\n"
  64. "\t Use voice file of this name from espeak-data/voices\n"
  65. "-w <wave file name>\n"
  66. "\t Write speech to this WAV file, rather than speaking it directly\n"
  67. "-b\t Input text encoding, 1=UTF8, 2=8 bit, 4=16 bit \n"
  68. "-m\t Interpret SSML markup, and ignore other < > tags\n"
  69. "-q\t Quiet, don't produce any speech (may be useful with -x)\n"
  70. "-x\t Write phoneme mnemonics to stdout\n"
  71. "-X\t Write phonemes mnemonics and translation trace to stdout\n"
  72. "-z\t No final sentence pause at the end of the text\n"
  73. "--compile=<voice name>\n"
  74. "\t Compile pronunciation rules and dictionary from the current\n"
  75. "\t directory. <voice name> specifies the language\n"
  76. "--compile-debug=<voice name>\n"
  77. "\t Compile pronunciation rules and dictionary from the current\n"
  78. "\t directory, including line numbers for use with -X.\n"
  79. "\t <voice name> specifies the language\n"
  80. "--compile-mbrola=<voice name>\n"
  81. "\t Compile an MBROLA voice\n"
  82. "--compile-intonations\n"
  83. "\t Compile the intonation data\n"
  84. "--compile-phonemes\n"
  85. "\t Compile the phoneme data\n"
  86. "--ipa Write phonemes to stdout using International Phonetic Alphabet\n"
  87. "--path=\"<path>\"\n"
  88. "\t Specifies the directory containing the espeak-data directory\n"
  89. "--pho Write mbrola phoneme data (.pho) to stdout or to the file in --phonout\n"
  90. "--phonout=\"<filename>\"\n"
  91. "\t Write phoneme output from -x -X --ipa and --pho to this file\n"
  92. "--punct=\"<characters>\"\n"
  93. "\t Speak the names of punctuation characters during speaking. If\n"
  94. "\t =<characters> is omitted, all punctuation is spoken.\n"
  95. "--sep=<character>\n"
  96. "\t Separate phonemes (from -x --ipa) with <character>.\n"
  97. "\t Default is space, z means ZWJN character.\n"
  98. "--split=<minutes>\n"
  99. "\t Starts a new WAV file every <minutes>. Used with -w\n"
  100. "--stdout Write speech output to stdout\n"
  101. "--tie=<character>\n"
  102. "\t Use a tie character within multi-letter phoneme names.\n"
  103. "\t Default is U+361, z means ZWJ character.\n"
  104. "--version Shows version number and date, and location of espeak-data\n"
  105. "--voices=<language>\n"
  106. "\t List the available voices for the specified language.\n"
  107. "\t If <language> is omitted, then list all voices.\n";
  108. int samplerate;
  109. int quiet = 0;
  110. unsigned int samples_total = 0;
  111. unsigned int samples_split = 0;
  112. unsigned int samples_split_seconds = 0;
  113. unsigned int wavefile_count = 0;
  114. FILE *f_wavfile = NULL;
  115. char filetype[5];
  116. char wavefile[200];
  117. void DisplayVoices(FILE *f_out, char *language)
  118. {
  119. int ix;
  120. const char *p;
  121. int len;
  122. int count;
  123. int c;
  124. size_t j;
  125. const espeak_VOICE *v;
  126. const char *lang_name;
  127. char age_buf[12];
  128. char buf[80];
  129. const espeak_VOICE **voices;
  130. espeak_VOICE voice_select;
  131. static char genders[4] = { '-', 'M', 'F', '-' };
  132. if ((language != NULL) && (language[0] != 0)) {
  133. // display only voices for the specified language, in order of priority
  134. voice_select.languages = language;
  135. voice_select.age = 0;
  136. voice_select.gender = 0;
  137. voice_select.name = NULL;
  138. voices = espeak_ListVoices(&voice_select);
  139. } else
  140. voices = espeak_ListVoices(NULL);
  141. fprintf(f_out, "Pty Language Age/Gender VoiceName File Other Languages\n");
  142. for (ix = 0; (v = voices[ix]) != NULL; ix++) {
  143. count = 0;
  144. p = v->languages;
  145. while (*p != 0) {
  146. len = strlen(p+1);
  147. lang_name = p+1;
  148. if (v->age == 0)
  149. strcpy(age_buf, " --");
  150. else
  151. sprintf(age_buf, "%3d", v->age);
  152. if (count == 0) {
  153. for (j = 0; j < sizeof(buf); j++) {
  154. // replace spaces in the name
  155. if ((c = v->name[j]) == ' ')
  156. c = '_';
  157. if ((buf[j] = c) == 0)
  158. break;
  159. }
  160. fprintf(f_out, "%2d %-15s%s/%c %-18s %-20s ",
  161. p[0], lang_name, age_buf, genders[v->gender], buf, v->identifier);
  162. } else
  163. fprintf(f_out, "(%s %d)", lang_name, p[0]);
  164. count++;
  165. p += len+2;
  166. }
  167. fputc('\n', f_out);
  168. }
  169. }
  170. static void Write4Bytes(FILE *f, int value)
  171. {
  172. // Write 4 bytes to a file, least significant first
  173. int ix;
  174. for (ix = 0; ix < 4; ix++) {
  175. fputc(value & 0xff, f);
  176. value = value >> 8;
  177. }
  178. }
  179. int OpenWavFile(char *path, int rate)
  180. {
  181. static unsigned char wave_hdr[44] = {
  182. 'R', 'I', 'F', 'F', 0x24, 0xf0, 0xff, 0x7f, 'W', 'A', 'V', 'E', 'f', 'm', 't', ' ',
  183. 0x10, 0, 0, 0, 1, 0, 1, 0, 9, 0x3d, 0, 0, 0x12, 0x7a, 0, 0,
  184. 2, 0, 0x10, 0, 'd', 'a', 't', 'a', 0x00, 0xf0, 0xff, 0x7f
  185. };
  186. if (path == NULL)
  187. return 2;
  188. while (isspace(*path)) path++;
  189. f_wavfile = NULL;
  190. if (path[0] != 0) {
  191. if (strcmp(path, "stdout") == 0) {
  192. #ifdef PLATFORM_WINDOWS
  193. // prevent Windows adding 0x0d before 0x0a bytes
  194. _setmode(_fileno(stdout), _O_BINARY);
  195. #endif
  196. f_wavfile = stdout;
  197. } else
  198. f_wavfile = fopen(path, "wb");
  199. }
  200. if (f_wavfile == NULL) {
  201. fprintf(stderr, "Can't write to: '%s'\n", path);
  202. return 1;
  203. }
  204. fwrite(wave_hdr, 1, 24, f_wavfile);
  205. Write4Bytes(f_wavfile, rate);
  206. Write4Bytes(f_wavfile, rate * 2);
  207. fwrite(&wave_hdr[32], 1, 12, f_wavfile);
  208. return 0;
  209. }
  210. static void CloseWavFile()
  211. {
  212. unsigned int pos;
  213. if ((f_wavfile == NULL) || (f_wavfile == stdout))
  214. return;
  215. fflush(f_wavfile);
  216. pos = ftell(f_wavfile);
  217. if (fseek(f_wavfile, 4, SEEK_SET) != -1)
  218. Write4Bytes(f_wavfile, pos - 8);
  219. if (fseek(f_wavfile, 40, SEEK_SET) != -1)
  220. Write4Bytes(f_wavfile, pos - 44);
  221. fclose(f_wavfile);
  222. f_wavfile = NULL;
  223. }
  224. static int SynthCallback(short *wav, int numsamples, espeak_EVENT *events)
  225. {
  226. char fname[210];
  227. if (quiet) return 0; // -q quiet mode
  228. if (wav == NULL) {
  229. CloseWavFile();
  230. return 0;
  231. }
  232. while (events->type != 0) {
  233. if (events->type == espeakEVENT_SAMPLERATE) {
  234. samplerate = events->id.number;
  235. samples_split = samples_split_seconds * samplerate;
  236. } else if (events->type == espeakEVENT_SENTENCE) {
  237. // start a new WAV file when the limit is reached, at this sentence boundary
  238. if ((samples_split > 0) && (samples_total > samples_split)) {
  239. CloseWavFile();
  240. samples_total = 0;
  241. wavefile_count++;
  242. }
  243. }
  244. events++;
  245. }
  246. if (f_wavfile == NULL) {
  247. if (samples_split > 0) {
  248. sprintf(fname, "%s_%.2d%s", wavefile, wavefile_count+1, filetype);
  249. if (OpenWavFile(fname, samplerate) != 0)
  250. return 1;
  251. } else if (OpenWavFile(wavefile, samplerate) != 0)
  252. return 1;
  253. }
  254. if (numsamples > 0) {
  255. samples_total += numsamples;
  256. fwrite(wav, numsamples*2, 1, f_wavfile);
  257. }
  258. return 0;
  259. }
  260. static void PrintVersion()
  261. {
  262. const char *version;
  263. const char *path_data;
  264. espeak_Initialize(AUDIO_OUTPUT_SYNCHRONOUS, 0, NULL, espeakINITIALIZE_DONT_EXIT);
  265. version = espeak_Info(&path_data);
  266. printf("eSpeak text-to-speech: %s Data at: %s\n", version, path_data);
  267. }
  268. int main(int argc, char **argv)
  269. {
  270. static struct option long_options[] = {
  271. { "help", no_argument, 0, 'h' },
  272. { "stdin", no_argument, 0, 0x100 },
  273. { "compile-debug", optional_argument, 0, 0x101 },
  274. { "compile", optional_argument, 0, 0x102 },
  275. { "punct", optional_argument, 0, 0x103 },
  276. { "voices", optional_argument, 0, 0x104 },
  277. { "stdout", no_argument, 0, 0x105 },
  278. { "split", optional_argument, 0, 0x106 },
  279. { "path", required_argument, 0, 0x107 },
  280. { "phonout", required_argument, 0, 0x108 },
  281. { "pho", no_argument, 0, 0x109 },
  282. { "ipa", optional_argument, 0, 0x10a },
  283. { "version", no_argument, 0, 0x10b },
  284. { "sep", optional_argument, 0, 0x10c },
  285. { "tie", optional_argument, 0, 0x10d },
  286. { "compile-mbrola", optional_argument, 0, 0x10e },
  287. { "compile-intonations", no_argument, 0, 0x10f },
  288. { "compile-phonemes", no_argument, 0, 0x110 },
  289. { 0, 0, 0, 0 }
  290. };
  291. FILE *f_text = NULL;
  292. char *p_text = NULL;
  293. FILE *f_phonemes_out = stdout;
  294. char *data_path = NULL; // use default path for espeak-data
  295. int option_index = 0;
  296. int c;
  297. int ix;
  298. char *optarg2;
  299. int value;
  300. int flag_stdin = 0;
  301. int flag_compile = 0;
  302. int filesize = 0;
  303. int synth_flags = espeakCHARS_AUTO | espeakPHONEMES | espeakENDPAUSE;
  304. int volume = -1;
  305. int speed = -1;
  306. int pitch = -1;
  307. int wordgap = -1;
  308. int option_capitals = -1;
  309. int option_punctuation = -1;
  310. int phonemes_separator = 0;
  311. int phoneme_options = 0;
  312. int option_linelength = 0;
  313. int option_waveout = 0;
  314. espeak_VOICE voice_select;
  315. char filename[200];
  316. char voicename[40];
  317. char devicename[200];
  318. #define N_PUNCTLIST 100
  319. wchar_t option_punctlist[N_PUNCTLIST];
  320. voicename[0] = 0;
  321. wavefile[0] = 0;
  322. filename[0] = 0;
  323. devicename[0] = 0;
  324. option_punctlist[0] = 0;
  325. while (true) {
  326. c = getopt_long(argc, argv, "a:b:d:f:g:hk:l:mp:qs:v:w:xXz",
  327. long_options, &option_index);
  328. // Detect the end of the options.
  329. if (c == -1)
  330. break;
  331. optarg2 = optarg;
  332. switch (c)
  333. {
  334. case 'b':
  335. // input character encoding, 8bit, 16bit, UTF8
  336. if ((sscanf(optarg2, "%d", &value) == 1) && (value <= 4))
  337. synth_flags |= value;
  338. else
  339. synth_flags |= espeakCHARS_8BIT;
  340. break;
  341. case 'd':
  342. strncpy0(devicename, optarg2, sizeof(devicename));
  343. break;
  344. case 'h':
  345. printf("\n");
  346. PrintVersion();
  347. printf("%s", help_text);
  348. return 0;
  349. case 'k':
  350. option_capitals = atoi(optarg2);
  351. break;
  352. case 'x':
  353. phoneme_options |= espeakPHONEMES_SHOW;
  354. break;
  355. case 'X':
  356. phoneme_options |= espeakPHONEMES_TRACE;
  357. break;
  358. case 'm':
  359. synth_flags |= espeakSSML;
  360. break;
  361. case 'p':
  362. pitch = atoi(optarg2);
  363. break;
  364. case 'q':
  365. quiet = 1;
  366. break;
  367. case 'f':
  368. strncpy0(filename, optarg2, sizeof(filename));
  369. break;
  370. case 'l':
  371. option_linelength = atoi(optarg2);
  372. break;
  373. case 'a':
  374. volume = atoi(optarg2);
  375. break;
  376. case 's':
  377. speed = atoi(optarg2);
  378. break;
  379. case 'g':
  380. wordgap = atoi(optarg2);
  381. break;
  382. case 'v':
  383. strncpy0(voicename, optarg2, sizeof(voicename));
  384. break;
  385. case 'w':
  386. option_waveout = 1;
  387. strncpy0(wavefile, optarg2, sizeof(filename));
  388. break;
  389. case 'z': // remove pause from the end of a sentence
  390. synth_flags &= ~espeakENDPAUSE;
  391. break;
  392. case 0x100: // --stdin
  393. flag_stdin = 1;
  394. break;
  395. case 0x105: // --stdout
  396. option_waveout = 1;
  397. strcpy(wavefile, "stdout");
  398. break;
  399. case 0x101: // --compile-debug
  400. case 0x102: // --compile
  401. strncpy0(voicename, optarg2, sizeof(voicename));
  402. flag_compile = c;
  403. quiet = 1;
  404. break;
  405. case 0x103: // --punct
  406. option_punctuation = 1;
  407. if (optarg2 != NULL) {
  408. ix = 0;
  409. while ((ix < N_PUNCTLIST) && ((option_punctlist[ix] = optarg2[ix]) != 0)) ix++;
  410. option_punctlist[N_PUNCTLIST-1] = 0;
  411. option_punctuation = 2;
  412. }
  413. break;
  414. case 0x104: // --voices
  415. espeak_Initialize(AUDIO_OUTPUT_SYNCHRONOUS, 0, data_path, 0);
  416. DisplayVoices(stdout, optarg2);
  417. exit(0);
  418. case 0x106: // -- split
  419. if (optarg2 == NULL)
  420. samples_split_seconds = 30 * 60; // default 30 minutes
  421. else
  422. samples_split_seconds = atoi(optarg2) * 60;
  423. break;
  424. case 0x107: // --path
  425. data_path = optarg2;
  426. break;
  427. case 0x108: // --phonout
  428. if ((f_phonemes_out = fopen(optarg2, "w")) == NULL)
  429. fprintf(stderr, "Can't write to: %s\n", optarg2);
  430. break;
  431. case 0x109: // --pho
  432. phoneme_options |= espeakPHONEMES_MBROLA;
  433. break;
  434. case 0x10a: // --ipa
  435. phoneme_options |= espeakPHONEMES_IPA;
  436. if (optarg2 != NULL) {
  437. // deprecated and obsolete
  438. switch (atoi(optarg2))
  439. {
  440. case 1:
  441. phonemes_separator = '_';
  442. break;
  443. case 2:
  444. phonemes_separator = 0x0361;
  445. phoneme_options |= espeakPHONEMES_TIE;
  446. break;
  447. case 3:
  448. phonemes_separator = 0x200d; // ZWJ
  449. phoneme_options |= espeakPHONEMES_TIE;
  450. break;
  451. }
  452. }
  453. break;
  454. case 0x10b: // --version
  455. PrintVersion();
  456. exit(0);
  457. case 0x10c: // --sep
  458. phoneme_options |= espeakPHONEMES_SHOW;
  459. if (optarg2 == 0)
  460. phonemes_separator = ' ';
  461. else
  462. utf8_in(&phonemes_separator, optarg2);
  463. if (phonemes_separator == 'z')
  464. phonemes_separator = 0x200c; // ZWNJ
  465. break;
  466. case 0x10d: // --tie
  467. phoneme_options |= (espeakPHONEMES_SHOW | espeakPHONEMES_TIE);
  468. if (optarg2 == 0)
  469. phonemes_separator = 0x0361; // default: combining-double-inverted-breve
  470. else
  471. utf8_in(&phonemes_separator, optarg2);
  472. if (phonemes_separator == 'z')
  473. phonemes_separator = 0x200d; // ZWJ
  474. break;
  475. case 0x10e: // --compile-mbrola
  476. {
  477. espeak_ng_InitializePath(data_path);
  478. espeak_ng_ERROR_CONTEXT context = NULL;
  479. espeak_ng_STATUS result = espeak_ng_CompileMbrolaVoice(optarg2, stdout, &context);
  480. if (result != ENS_OK) {
  481. espeak_ng_PrintStatusCodeMessage(result, stderr, context);
  482. espeak_ng_ClearErrorContext(&context);
  483. return EXIT_FAILURE;
  484. }
  485. return EXIT_SUCCESS;
  486. }
  487. case 0x10f: // --compile-intonations
  488. {
  489. espeak_ng_InitializePath(data_path);
  490. espeak_ng_ERROR_CONTEXT context = NULL;
  491. espeak_ng_STATUS result = espeak_ng_CompileIntonation(stdout, &context);
  492. if (result != ENS_OK) {
  493. espeak_ng_PrintStatusCodeMessage(result, stderr, context);
  494. espeak_ng_ClearErrorContext(&context);
  495. return EXIT_FAILURE;
  496. }
  497. return EXIT_SUCCESS;
  498. }
  499. case 0x110: // --compile-phonemes
  500. {
  501. espeak_ng_InitializePath(data_path);
  502. espeak_ng_ERROR_CONTEXT context = NULL;
  503. espeak_ng_STATUS result = espeak_ng_CompilePhonemeData(22050, stdout, &context);
  504. if (result != ENS_OK) {
  505. espeak_ng_PrintStatusCodeMessage(result, stderr, context);
  506. espeak_ng_ClearErrorContext(&context);
  507. return EXIT_FAILURE;
  508. }
  509. return EXIT_SUCCESS;
  510. }
  511. default:
  512. exit(0);
  513. }
  514. }
  515. espeak_ng_InitializePath(data_path);
  516. espeak_ng_ERROR_CONTEXT context = NULL;
  517. espeak_ng_STATUS result = espeak_ng_Initialize(&context);
  518. if (result != ENS_OK) {
  519. espeak_ng_PrintStatusCodeMessage(result, stderr, context);
  520. espeak_ng_ClearErrorContext(&context);
  521. exit(1);
  522. }
  523. if (option_waveout || quiet) {
  524. // writing to a file (or no output), we can use synchronous mode
  525. result = espeak_ng_InitializeOutput(ENOUTPUT_MODE_SYNCHRONOUS, 0, devicename[0] ? devicename : NULL);
  526. samplerate = espeak_ng_GetSampleRate();
  527. samples_split = samplerate * samples_split_seconds;
  528. espeak_SetSynthCallback(SynthCallback);
  529. if (samples_split) {
  530. char *extn;
  531. extn = strrchr(wavefile, '.');
  532. if ((extn != NULL) && ((wavefile + strlen(wavefile) - extn) <= 4)) {
  533. strcpy(filetype, extn);
  534. *extn = 0;
  535. }
  536. }
  537. } else {
  538. // play the sound output
  539. result = espeak_ng_InitializeOutput(PLAYBACK_MODE, 0, devicename[0] ? devicename : NULL);
  540. samplerate = espeak_ng_GetSampleRate();
  541. }
  542. if (result != ENS_OK) {
  543. espeak_ng_PrintStatusCodeMessage(result, stderr, NULL);
  544. exit(EXIT_FAILURE);
  545. }
  546. if (voicename[0] == 0)
  547. strcpy(voicename, "default");
  548. result = espeak_ng_SetVoiceByName(voicename);
  549. if (result != ENS_OK) {
  550. memset(&voice_select, 0, sizeof(voice_select));
  551. voice_select.languages = voicename;
  552. result = espeak_ng_SetVoiceByProperties(&voice_select);
  553. if (result != ENS_OK) {
  554. espeak_ng_PrintStatusCodeMessage(result, stderr, NULL);
  555. exit(EXIT_FAILURE);
  556. }
  557. }
  558. if (flag_compile) {
  559. // This must be done after the voice is set
  560. espeak_ng_ERROR_CONTEXT context = NULL;
  561. espeak_ng_STATUS result = espeak_ng_CompileDictionary("", NULL, stderr, flag_compile & 0x1, &context);
  562. if (result != ENS_OK) {
  563. espeak_ng_PrintStatusCodeMessage(result, stderr, context);
  564. espeak_ng_ClearErrorContext(&context);
  565. return EXIT_FAILURE;
  566. }
  567. return EXIT_SUCCESS;
  568. }
  569. // set any non-default values of parameters. This must be done after espeak_Initialize()
  570. if (speed > 0)
  571. espeak_SetParameter(espeakRATE, speed, 0);
  572. if (volume >= 0)
  573. espeak_SetParameter(espeakVOLUME, volume, 0);
  574. if (pitch >= 0)
  575. espeak_SetParameter(espeakPITCH, pitch, 0);
  576. if (option_capitals >= 0)
  577. espeak_SetParameter(espeakCAPITALS, option_capitals, 0);
  578. if (option_punctuation >= 0)
  579. espeak_SetParameter(espeakPUNCTUATION, option_punctuation, 0);
  580. if (wordgap >= 0)
  581. espeak_SetParameter(espeakWORDGAP, wordgap, 0);
  582. if (option_linelength > 0)
  583. espeak_SetParameter(espeakLINELENGTH, option_linelength, 0);
  584. if (option_punctuation == 2)
  585. espeak_SetPunctuationList(option_punctlist);
  586. espeak_SetPhonemeTrace(phoneme_options | (phonemes_separator << 8), f_phonemes_out);
  587. if (filename[0] == 0) {
  588. if ((optind < argc) && (flag_stdin == 0)) {
  589. // there's a non-option parameter, and no -f or --stdin
  590. // use it as text
  591. p_text = argv[optind];
  592. } else {
  593. f_text = stdin;
  594. if (flag_stdin == 0)
  595. flag_stdin = 2;
  596. }
  597. } else {
  598. filesize = GetFileLength(filename);
  599. f_text = fopen(filename, "r");
  600. if (f_text == NULL) {
  601. fprintf(stderr, "Failed to read file '%s'\n", filename);
  602. exit(EXIT_FAILURE);
  603. }
  604. }
  605. if (p_text != NULL) {
  606. int size;
  607. size = strlen(p_text);
  608. espeak_Synth(p_text, size+1, 0, POS_CHARACTER, 0, synth_flags, NULL, NULL);
  609. } else if (flag_stdin) {
  610. int max = 1000;
  611. if ((p_text = (char *)malloc(max)) == NULL) {
  612. espeak_ng_PrintStatusCodeMessage(ENOMEM, stderr, NULL);
  613. exit(EXIT_FAILURE);
  614. }
  615. if (flag_stdin == 2) {
  616. // line by line input on stdin
  617. while (fgets(p_text, max, stdin) != NULL) {
  618. p_text[max-1] = 0;
  619. espeak_Synth(p_text, max, 0, POS_CHARACTER, 0, synth_flags, NULL, NULL);
  620. }
  621. } else {
  622. // bulk input on stdin
  623. ix = 0;
  624. while (true) {
  625. if ((c = fgetc(stdin)) == EOF)
  626. break;
  627. p_text[ix++] = (char)c;
  628. if (ix >= (max-1)) {
  629. max += 1000;
  630. char *new_text = (char *)realloc(p_text, max);
  631. if (new_text == NULL) {
  632. free(p_text);
  633. espeak_ng_PrintStatusCodeMessage(ENOMEM, stderr, NULL);
  634. exit(EXIT_FAILURE);
  635. }
  636. }
  637. }
  638. if (ix > 0) {
  639. p_text[ix-1] = 0;
  640. espeak_Synth(p_text, ix+1, 0, POS_CHARACTER, 0, synth_flags, NULL, NULL);
  641. }
  642. }
  643. } else if (f_text != NULL) {
  644. if ((p_text = (char *)malloc(filesize+1)) == NULL) {
  645. espeak_ng_PrintStatusCodeMessage(ENOMEM, stderr, NULL);
  646. exit(EXIT_FAILURE);
  647. }
  648. fread(p_text, 1, filesize, f_text);
  649. p_text[filesize] = 0;
  650. espeak_Synth(p_text, filesize+1, 0, POS_CHARACTER, 0, synth_flags, NULL, NULL);
  651. fclose(f_text);
  652. }
  653. result = espeak_ng_Synchronize();
  654. if (result != ENS_OK) {
  655. espeak_ng_PrintStatusCodeMessage(result, stderr, NULL);
  656. exit(EXIT_FAILURE);
  657. }
  658. if (f_phonemes_out != stdout)
  659. fclose(f_phonemes_out);
  660. return 0;
  661. }