1 /* 2 * --------------------------------------------------------------------------- 3 * Recognizer.java 4 * 5 * Copyright 2007 Nuance Communciations, Inc. 6 * 7 * Licensed under the Apache License, Version 2.0 (the 'License'); you may not 8 * use this file except in compliance with the License. 9 * 10 * You may obtain a copy of the License at 11 * http://www.apache.org/licenses/LICENSE-2.0 12 * 13 * Unless required by applicable law or agreed to in writing, software 14 * distributed under the License is distributed on an 'AS IS' BASIS, WITHOUT 15 * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the 16 * License for the specific language governing permissions and limitations under 17 * the License. 18 * 19 * --------------------------------------------------------------------------- 20 */ 21 22 23 package android.speech.srec; 24 25 import android.util.Log; 26 27 import java.io.File; 28 import java.io.InputStream; 29 import java.io.IOException; 30 import java.util.Locale; 31 32 /** 33 * Simple, synchronous speech recognizer, using the Nuance SREC package. 34 * Usages proceeds as follows: 35 * 36 * <ul> 37 * <li>Create a <code>Recognizer</code>. 38 * <li>Create a <code>Recognizer.Grammar</code>. 39 * <li>Setup the <code>Recognizer.Grammar</code>. 40 * <li>Reset the <code>Recognizer.Grammar</code> slots, if needed. 41 * <li>Fill the <code>Recognizer.Grammar</code> slots, if needed. 42 * <li>Compile the <code>Recognizer.Grammar</code>, if needed. 43 * <li>Save the filled <code>Recognizer.Grammar</code>, if needed. 44 * <li>Start the <code>Recognizer</code>. 45 * <li>Loop over <code>advance</code> and <code>putAudio</code> until recognition complete. 46 * <li>Fetch and process results, or notify of failure. 47 * <li>Stop the <code>Recognizer</code>. 48 * <li>Destroy the <code>Recognizer</code>. 49 * </ul> 50 * 51 * <p>Below is example code</p> 52 * 53 * <pre class="prettyprint"> 54 * 55 * // create and start audio input 56 * InputStream audio = new MicrophoneInputStream(11025, 11025*5); 57 * // create a Recognizer 58 * String cdir = Recognizer.getConfigDir(null); 59 * Recognizer recognizer = new Recognizer(cdir + "/baseline11k.par"); 60 * // create and load a Grammar 61 * Recognizer.Grammar grammar = recognizer.new Grammar(cdir + "/grammars/VoiceDialer.g2g"); 62 * // setup the Grammar to work with the Recognizer 63 * grammar.setupRecognizer(); 64 * // fill the Grammar slots with names and save, if required 65 * grammar.resetAllSlots(); 66 * for (String name : names) grammar.addWordToSlot("@Names", name, null, 1, "V=1"); 67 * grammar.compile(); 68 * grammar.save(".../foo.g2g"); 69 * // start the Recognizer 70 * recognizer.start(); 71 * // loop over Recognizer events 72 * while (true) { 73 * switch (recognizer.advance()) { 74 * case Recognizer.EVENT_INCOMPLETE: 75 * case Recognizer.EVENT_STARTED: 76 * case Recognizer.EVENT_START_OF_VOICING: 77 * case Recognizer.EVENT_END_OF_VOICING: 78 * // let the Recognizer continue to run 79 * continue; 80 * case Recognizer.EVENT_RECOGNITION_RESULT: 81 * // success, so fetch results here! 82 * for (int i = 0; i < recognizer.getResultCount(); i++) { 83 * String result = recognizer.getResult(i, Recognizer.KEY_LITERAL); 84 * } 85 * break; 86 * case Recognizer.EVENT_NEED_MORE_AUDIO: 87 * // put more audio in the Recognizer 88 * recognizer.putAudio(audio); 89 * continue; 90 * default: 91 * notifyFailure(); 92 * break; 93 * } 94 * break; 95 * } 96 * // stop the Recognizer 97 * recognizer.stop(); 98 * // destroy the Recognizer 99 * recognizer.destroy(); 100 * // stop the audio device 101 * audio.close(); 102 * 103 * </pre> 104 */ 105 public final class Recognizer { 106 static { 107 System.loadLibrary("srec_jni"); 108 } 109 110 private static String TAG = "Recognizer"; 111 112 /** 113 * Result key corresponding to confidence score. 114 */ 115 public static final String KEY_CONFIDENCE = "conf"; 116 117 /** 118 * Result key corresponding to literal text. 119 */ 120 public static final String KEY_LITERAL = "literal"; 121 122 /** 123 * Result key corresponding to semantic meaning text. 124 */ 125 public static final String KEY_MEANING = "meaning"; 126 127 // handle to SR_Vocabulary object 128 private int mVocabulary = 0; 129 130 // handle to SR_Recognizer object 131 private int mRecognizer = 0; 132 133 // Grammar currently associated with Recognizer via SR_GrammarSetupRecognizer 134 private Grammar mActiveGrammar = null; 135 136 /** 137 * Get the pathname of the SREC configuration directory corresponding to the 138 * language indicated by the Locale. 139 * This directory contains dictionaries, speech models, 140 * configuration files, and other data needed by the Recognizer. 141 * @param locale <code>Locale</code> corresponding to the desired language, 142 * or null for default, currently <code>Locale.US</code>. 143 * @return Pathname of the configuration directory. 144 */ 145 public static String getConfigDir(Locale locale) { 146 if (locale == null) locale = Locale.US; 147 String dir = "/system/usr/srec/config/" + 148 locale.toString().replace('_', '.').toLowerCase(Locale.ROOT); 149 if ((new File(dir)).isDirectory()) return dir; 150 return null; 151 } 152 153 /** 154 * Create an instance of a SREC speech recognizer. 155 * 156 * @param configFile pathname of the baseline*.par configuration file, 157 * which in turn contains references to dictionaries, speech models, 158 * and other data needed to configure and operate the recognizer. 159 * A separate config file is needed for each audio sample rate. 160 * Two files, baseline11k.par and baseline8k.par, which correspond to 161 * 11025 and 8000 hz, are present in the directory indicated by 162 * {@link #getConfigDir}. 163 * @throws IOException 164 */ 165 public Recognizer(String configFile) throws IOException { 166 PMemInit(); 167 SR_SessionCreate(configFile); 168 mRecognizer = SR_RecognizerCreate(); 169 SR_RecognizerSetup(mRecognizer); 170 mVocabulary = SR_VocabularyLoad(); 171 } 172 173 /** 174 * Represents a grammar loaded into the Recognizer. 175 */ 176 public class Grammar { 177 private int mGrammar = 0; 178 179 /** 180 * Create a <code>Grammar</code> instance. 181 * @param g2gFileName pathname of g2g file. 182 */ 183 public Grammar(String g2gFileName) throws IOException { 184 mGrammar = SR_GrammarLoad(g2gFileName); 185 SR_GrammarSetupVocabulary(mGrammar, mVocabulary); 186 } 187 188 /** 189 * Reset all slots. 190 */ 191 public void resetAllSlots() { 192 SR_GrammarResetAllSlots(mGrammar); 193 } 194 195 /** 196 * Add a word to a slot. 197 * 198 * @param slot slot name. 199 * @param word word to insert. 200 * @param pron pronunciation, or null to derive from word. 201 * @param weight weight to give the word. One is normal, 50 is low. 202 * @param tag semantic meaning tag string. 203 */ 204 public void addWordToSlot(String slot, String word, String pron, int weight, String tag) { 205 SR_GrammarAddWordToSlot(mGrammar, slot, word, pron, weight, tag); 206 } 207 208 /** 209 * Compile all slots. 210 */ 211 public void compile() { 212 SR_GrammarCompile(mGrammar); 213 } 214 215 /** 216 * Setup <code>Grammar</code> with <code>Recognizer</code>. 217 */ 218 public void setupRecognizer() { 219 SR_GrammarSetupRecognizer(mGrammar, mRecognizer); 220 mActiveGrammar = this; 221 } 222 223 /** 224 * Save <code>Grammar</code> to g2g file. 225 * 226 * @param g2gFileName 227 * @throws IOException 228 */ 229 public void save(String g2gFileName) throws IOException { 230 SR_GrammarSave(mGrammar, g2gFileName); 231 } 232 233 /** 234 * Release resources associated with this <code>Grammar</code>. 235 */ 236 public void destroy() { 237 // TODO: need to do cleanup and disassociation with Recognizer 238 if (mGrammar != 0) { 239 SR_GrammarDestroy(mGrammar); 240 mGrammar = 0; 241 } 242 } 243 244 /** 245 * Clean up resources. 246 */ 247 protected void finalize() { 248 if (mGrammar != 0) { 249 destroy(); 250 throw new IllegalStateException("someone forgot to destroy Grammar"); 251 } 252 } 253 } 254 255 /** 256 * Start recognition 257 */ 258 public void start() { 259 // TODO: shouldn't be here? 260 SR_RecognizerActivateRule(mRecognizer, mActiveGrammar.mGrammar, "trash", 1); 261 SR_RecognizerStart(mRecognizer); 262 } 263 264 /** 265 * Process some audio and return the current status. 266 * @return recognition event, one of: 267 * <ul> 268 * <li><code>EVENT_INVALID</code> 269 * <li><code>EVENT_NO_MATCH</code> 270 * <li><code>EVENT_INCOMPLETE</code> 271 * <li><code>EVENT_STARTED</code> 272 * <li><code>EVENT_STOPPED</code> 273 * <li><code>EVENT_START_OF_VOICING</code> 274 * <li><code>EVENT_END_OF_VOICING</code> 275 * <li><code>EVENT_SPOKE_TOO_SOON</code> 276 * <li><code>EVENT_RECOGNITION_RESULT</code> 277 * <li><code>EVENT_START_OF_UTTERANCE_TIMEOUT</code> 278 * <li><code>EVENT_RECOGNITION_TIMEOUT</code> 279 * <li><code>EVENT_NEED_MORE_AUDIO</code> 280 * <li><code>EVENT_MAX_SPEECH</code> 281 * </ul> 282 */ 283 public int advance() { 284 return SR_RecognizerAdvance(mRecognizer); 285 } 286 287 /** 288 * Put audio samples into the <code>Recognizer</code>. 289 * @param buf holds the audio samples. 290 * @param offset offset of the first sample. 291 * @param length number of bytes containing samples. 292 * @param isLast indicates no more audio data, normally false. 293 * @return number of bytes accepted. 294 */ 295 public int putAudio(byte[] buf, int offset, int length, boolean isLast) { 296 return SR_RecognizerPutAudio(mRecognizer, buf, offset, length, isLast); 297 } 298 299 /** 300 * Read audio samples from an <code>InputStream</code> and put them in the 301 * <code>Recognizer</code>. 302 * @param audio <code>InputStream</code> containing PCM audio samples. 303 */ 304 public void putAudio(InputStream audio) throws IOException { 305 // make sure the audio buffer is allocated 306 if (mPutAudioBuffer == null) mPutAudioBuffer = new byte[512]; 307 // read some data 308 int nbytes = audio.read(mPutAudioBuffer); 309 // eof, so signal Recognizer 310 if (nbytes == -1) { 311 SR_RecognizerPutAudio(mRecognizer, mPutAudioBuffer, 0, 0, true); 312 } 313 // put it into the Recognizer 314 else if (nbytes != SR_RecognizerPutAudio(mRecognizer, mPutAudioBuffer, 0, nbytes, false)) { 315 throw new IOException("SR_RecognizerPutAudio failed nbytes=" + nbytes); 316 } 317 } 318 319 // audio buffer for putAudio(InputStream) 320 private byte[] mPutAudioBuffer = null; 321 322 /** 323 * Get the number of recognition results. Must be called after 324 * <code>EVENT_RECOGNITION_RESULT</code> is returned by 325 * <code>advance</code>, but before <code>stop</code>. 326 * 327 * @return number of results in nbest list. 328 */ 329 public int getResultCount() { 330 return SR_RecognizerResultGetSize(mRecognizer); 331 } 332 333 /** 334 * Get a set of keys for the result. Must be called after 335 * <code>EVENT_RECOGNITION_RESULT</code> is returned by 336 * <code>advance</code>, but before <code>stop</code>. 337 * 338 * @param index index of result. 339 * @return array of keys. 340 */ 341 public String[] getResultKeys(int index) { 342 return SR_RecognizerResultGetKeyList(mRecognizer, index); 343 } 344 345 /** 346 * Get a result value. Must be called after 347 * <code>EVENT_RECOGNITION_RESULT</code> is returned by 348 * <code>advance</code>, but before <code>stop</code>. 349 * 350 * @param index index of the result. 351 * @param key key of the result. This is typically one of 352 * <code>KEY_CONFIDENCE</code>, <code>KEY_LITERAL</code>, or 353 * <code>KEY_MEANING</code>, but the user can also define their own keys 354 * in a grxml file, or in the <code>tag</code> slot of 355 * <code>Grammar.addWordToSlot</code>. 356 * @return the result. 357 */ 358 public String getResult(int index, String key) { 359 return SR_RecognizerResultGetValue(mRecognizer, index, key); 360 } 361 362 /** 363 * Stop the <code>Recognizer</code>. 364 */ 365 public void stop() { 366 SR_RecognizerStop(mRecognizer); 367 SR_RecognizerDeactivateRule(mRecognizer, mActiveGrammar.mGrammar, "trash"); 368 } 369 370 /** 371 * Reset the acoustic state vectorto it's default value. 372 * 373 * @hide 374 */ 375 public void resetAcousticState() { 376 SR_AcousticStateReset(mRecognizer); 377 } 378 379 /** 380 * Set the acoustic state vector. 381 * @param state String containing the acoustic state vector. 382 * 383 * @hide 384 */ 385 public void setAcousticState(String state) { 386 SR_AcousticStateSet(mRecognizer, state); 387 } 388 389 /** 390 * Get the acoustic state vector. 391 * @return String containing the acoustic state vector. 392 * 393 * @hide 394 */ 395 public String getAcousticState() { 396 return SR_AcousticStateGet(mRecognizer); 397 } 398 399 /** 400 * Clean up resources. 401 */ 402 public void destroy() { 403 try { 404 if (mVocabulary != 0) SR_VocabularyDestroy(mVocabulary); 405 } finally { 406 mVocabulary = 0; 407 try { 408 if (mRecognizer != 0) SR_RecognizerUnsetup(mRecognizer); 409 } finally { 410 try { 411 if (mRecognizer != 0) SR_RecognizerDestroy(mRecognizer); 412 } finally { 413 mRecognizer = 0; 414 try { 415 SR_SessionDestroy(); 416 } finally { 417 PMemShutdown(); 418 } 419 } 420 } 421 } 422 } 423 424 /** 425 * Clean up resources. 426 */ 427 protected void finalize() throws Throwable { 428 if (mVocabulary != 0 || mRecognizer != 0) { 429 destroy(); 430 throw new IllegalStateException("someone forgot to destroy Recognizer"); 431 } 432 } 433 434 /* an example session captured, for reference 435 void doall() { 436 if (PMemInit ( ) 437 || lhs_audioinOpen ( WAVE_MAPPER, SREC_TEST_DEFAULT_AUDIO_FREQUENCY, &audio_in_handle ) 438 || srec_test_init_application_data ( &applicationData, argc, argv ) 439 || SR_SessionCreate ( "/system/usr/srec/config/en.us/baseline11k.par" ) 440 || SR_RecognizerCreate ( &applicationData.recognizer ) 441 || SR_RecognizerSetup ( applicationData.recognizer) 442 || ESR_SessionGetLCHAR ( L("cmdline.vocabulary"), filename, &flen ) 443 || SR_VocabularyLoad ( filename, &applicationData.vocabulary ) 444 || SR_VocabularyGetLanguage ( applicationData.vocabulary, &applicationData.locale ) 445 || (applicationData.nametag = NULL) 446 || SR_NametagsCreate ( &applicationData.nametags ) 447 || (LSTRCPY ( applicationData.grammars [0].grammar_path, "/system/usr/srec/config/en.us/grammars/VoiceDialer.g2g" ), 0) 448 || (LSTRCPY ( applicationData.grammars [0].grammarID, "BothTags" ), 0) 449 || (LSTRCPY ( applicationData.grammars [0].ruleName, "trash" ), 0) 450 || (applicationData.grammars [0].is_ve_grammar = ESR_FALSE, 0) 451 || SR_GrammarLoad (applicationData.grammars [0].grammar_path, &applicationData.grammars [applicationData.grammarCount].grammar ) 452 || SR_GrammarSetupVocabulary ( applicationData.grammars [0].grammar, applicationData.vocabulary ) 453 || SR_GrammarSetupRecognizer( applicationData.grammars [0].grammar, applicationData.recognizer ) 454 || SR_GrammarSetDispatchFunction ( applicationData.grammars [0].grammar, L("myDSMCallback"), NULL, myDSMCallback ) 455 || (applicationData.grammarCount++, 0) 456 || SR_RecognizerActivateRule ( applicationData.recognizer, applicationData.grammars [0].grammar, 457 applicationData.grammars [0].ruleName, 1 ) 458 || (applicationData.active_grammar_num = 0, 0) 459 || lhs_audioinStart ( audio_in_handle ) 460 || SR_RecognizerStart ( applicationData.recognizer ) 461 || strl ( applicationData.grammars [0].grammar, &applicationData, audio_in_handle, &recognition_count ) 462 || SR_RecognizerStop ( applicationData.recognizer ) 463 || lhs_audioinStop ( audio_in_handle ) 464 || SR_RecognizerDeactivateRule ( applicationData.recognizer, applicationData.grammars [0].grammar, applicationData.grammars [0].ruleName ) 465 || (applicationData.active_grammar_num = -1, 0) 466 || SR_GrammarDestroy ( applicationData.grammars [0].grammar ) 467 || (applicationData.grammarCount--, 0) 468 || SR_NametagsDestroy ( applicationData.nametags ) 469 || (applicationData.nametags = NULL, 0) 470 || SR_VocabularyDestroy ( applicationData.vocabulary ) 471 || (applicationData.vocabulary = NULL) 472 || SR_RecognizerUnsetup ( applicationData.recognizer) // releases acoustic models 473 || SR_RecognizerDestroy ( applicationData.recognizer ) 474 || (applicationData.recognizer = NULL) 475 || SR_SessionDestroy ( ) 476 || srec_test_shutdown_application_data ( &applicationData ) 477 || lhs_audioinClose ( &audio_in_handle ) 478 || PMemShutdown ( ) 479 } 480 */ 481 482 483 // 484 // PMem native methods 485 // 486 private static native void PMemInit(); 487 private static native void PMemShutdown(); 488 489 490 // 491 // SR_Session native methods 492 // 493 private static native void SR_SessionCreate(String filename); 494 private static native void SR_SessionDestroy(); 495 496 497 // 498 // SR_Recognizer native methods 499 // 500 501 /** 502 * Reserved value. 503 */ 504 public final static int EVENT_INVALID = 0; 505 506 /** 507 * <code>Recognizer</code> could not find a match for the utterance. 508 */ 509 public final static int EVENT_NO_MATCH = 1; 510 511 /** 512 * <code>Recognizer</code> processed one frame of audio. 513 */ 514 public final static int EVENT_INCOMPLETE = 2; 515 516 /** 517 * <code>Recognizer</code> has just been started. 518 */ 519 public final static int EVENT_STARTED = 3; 520 521 /** 522 * <code>Recognizer</code> is stopped. 523 */ 524 public final static int EVENT_STOPPED = 4; 525 526 /** 527 * Beginning of speech detected. 528 */ 529 public final static int EVENT_START_OF_VOICING = 5; 530 531 /** 532 * End of speech detected. 533 */ 534 public final static int EVENT_END_OF_VOICING = 6; 535 536 /** 537 * Beginning of utterance occured too soon. 538 */ 539 public final static int EVENT_SPOKE_TOO_SOON = 7; 540 541 /** 542 * Recognition match detected. 543 */ 544 public final static int EVENT_RECOGNITION_RESULT = 8; 545 546 /** 547 * Timeout occured before beginning of utterance. 548 */ 549 public final static int EVENT_START_OF_UTTERANCE_TIMEOUT = 9; 550 551 /** 552 * Timeout occured before speech recognition could complete. 553 */ 554 public final static int EVENT_RECOGNITION_TIMEOUT = 10; 555 556 /** 557 * Not enough samples to process one frame. 558 */ 559 public final static int EVENT_NEED_MORE_AUDIO = 11; 560 561 /** 562 * More audio encountered than is allowed by 'swirec_max_speech_duration'. 563 */ 564 public final static int EVENT_MAX_SPEECH = 12; 565 566 /** 567 * Produce a displayable string from an <code>advance</code> event. 568 * @param event 569 * @return String representing the event. 570 */ 571 public static String eventToString(int event) { 572 switch (event) { 573 case EVENT_INVALID: 574 return "EVENT_INVALID"; 575 case EVENT_NO_MATCH: 576 return "EVENT_NO_MATCH"; 577 case EVENT_INCOMPLETE: 578 return "EVENT_INCOMPLETE"; 579 case EVENT_STARTED: 580 return "EVENT_STARTED"; 581 case EVENT_STOPPED: 582 return "EVENT_STOPPED"; 583 case EVENT_START_OF_VOICING: 584 return "EVENT_START_OF_VOICING"; 585 case EVENT_END_OF_VOICING: 586 return "EVENT_END_OF_VOICING"; 587 case EVENT_SPOKE_TOO_SOON: 588 return "EVENT_SPOKE_TOO_SOON"; 589 case EVENT_RECOGNITION_RESULT: 590 return "EVENT_RECOGNITION_RESULT"; 591 case EVENT_START_OF_UTTERANCE_TIMEOUT: 592 return "EVENT_START_OF_UTTERANCE_TIMEOUT"; 593 case EVENT_RECOGNITION_TIMEOUT: 594 return "EVENT_RECOGNITION_TIMEOUT"; 595 case EVENT_NEED_MORE_AUDIO: 596 return "EVENT_NEED_MORE_AUDIO"; 597 case EVENT_MAX_SPEECH: 598 return "EVENT_MAX_SPEECH"; 599 } 600 return "EVENT_" + event; 601 } 602 603 // 604 // SR_Recognizer methods 605 // 606 private static native void SR_RecognizerStart(int recognizer); 607 private static native void SR_RecognizerStop(int recognizer); 608 private static native int SR_RecognizerCreate(); 609 private static native void SR_RecognizerDestroy(int recognizer); 610 private static native void SR_RecognizerSetup(int recognizer); 611 private static native void SR_RecognizerUnsetup(int recognizer); 612 private static native boolean SR_RecognizerIsSetup(int recognizer); 613 private static native String SR_RecognizerGetParameter(int recognizer, String key); 614 private static native int SR_RecognizerGetSize_tParameter(int recognizer, String key); 615 private static native boolean SR_RecognizerGetBoolParameter(int recognizer, String key); 616 private static native void SR_RecognizerSetParameter(int recognizer, String key, String value); 617 private static native void SR_RecognizerSetSize_tParameter(int recognizer, 618 String key, int value); 619 private static native void SR_RecognizerSetBoolParameter(int recognizer, String key, 620 boolean value); 621 private static native void SR_RecognizerSetupRule(int recognizer, int grammar, 622 String ruleName); 623 private static native boolean SR_RecognizerHasSetupRules(int recognizer); 624 private static native void SR_RecognizerActivateRule(int recognizer, int grammar, 625 String ruleName, int weight); 626 private static native void SR_RecognizerDeactivateRule(int recognizer, int grammar, 627 String ruleName); 628 private static native void SR_RecognizerDeactivateAllRules(int recognizer); 629 private static native boolean SR_RecognizerIsActiveRule(int recognizer, int grammar, 630 String ruleName); 631 private static native boolean SR_RecognizerCheckGrammarConsistency(int recognizer, 632 int grammar); 633 private static native int SR_RecognizerPutAudio(int recognizer, byte[] buffer, int offset, 634 int length, boolean isLast); 635 private static native int SR_RecognizerAdvance(int recognizer); 636 // private static native void SR_RecognizerLoadUtterance(int recognizer, 637 // const LCHAR* filename); 638 // private static native void SR_RecognizerLoadWaveFile(int recognizer, 639 // const LCHAR* filename); 640 // private static native void SR_RecognizerSetLockFunction(int recognizer, 641 // SR_RecognizerLockFunction function, void* data); 642 private static native boolean SR_RecognizerIsSignalClipping(int recognizer); 643 private static native boolean SR_RecognizerIsSignalDCOffset(int recognizer); 644 private static native boolean SR_RecognizerIsSignalNoisy(int recognizer); 645 private static native boolean SR_RecognizerIsSignalTooQuiet(int recognizer); 646 private static native boolean SR_RecognizerIsSignalTooFewSamples(int recognizer); 647 private static native boolean SR_RecognizerIsSignalTooManySamples(int recognizer); 648 // private static native void SR_Recognizer_Change_Sample_Rate (size_t new_sample_rate); 649 650 651 // 652 // SR_AcousticState native methods 653 // 654 private static native void SR_AcousticStateReset(int recognizer); 655 private static native void SR_AcousticStateSet(int recognizer, String state); 656 private static native String SR_AcousticStateGet(int recognizer); 657 658 659 // 660 // SR_Grammar native methods 661 // 662 private static native void SR_GrammarCompile(int grammar); 663 private static native void SR_GrammarAddWordToSlot(int grammar, String slot, 664 String word, String pronunciation, int weight, String tag); 665 private static native void SR_GrammarResetAllSlots(int grammar); 666 // private static native void SR_GrammarAddNametagToSlot(int grammar, String slot, 667 // const struct SR_Nametag_t* nametag, int weight, String tag); 668 private static native void SR_GrammarSetupVocabulary(int grammar, int vocabulary); 669 // private static native void SR_GrammarSetupModels(int grammar, SR_AcousticModels* models); 670 private static native void SR_GrammarSetupRecognizer(int grammar, int recognizer); 671 private static native void SR_GrammarUnsetupRecognizer(int grammar); 672 // private static native void SR_GrammarGetModels(int grammar,SR_AcousticModels** models); 673 private static native int SR_GrammarCreate(); 674 private static native void SR_GrammarDestroy(int grammar); 675 private static native int SR_GrammarLoad(String filename); 676 private static native void SR_GrammarSave(int grammar, String filename); 677 // private static native void SR_GrammarSetDispatchFunction(int grammar, 678 // const LCHAR* name, void* userData, SR_GrammarDispatchFunction function); 679 // private static native void SR_GrammarSetParameter(int grammar, const 680 // LCHAR* key, void* value); 681 // private static native void SR_GrammarSetSize_tParameter(int grammar, 682 // const LCHAR* key, size_t value); 683 // private static native void SR_GrammarGetParameter(int grammar, const 684 // LCHAR* key, void** value); 685 // private static native void SR_GrammarGetSize_tParameter(int grammar, 686 // const LCHAR* key, size_t* value); 687 // private static native void SR_GrammarCheckParse(int grammar, const LCHAR* 688 // transcription, SR_SemanticResult** result, size_t* resultCount); 689 private static native void SR_GrammarAllowOnly(int grammar, String transcription); 690 private static native void SR_GrammarAllowAll(int grammar); 691 692 693 // 694 // SR_Vocabulary native methods 695 // 696 // private static native int SR_VocabularyCreate(); 697 private static native int SR_VocabularyLoad(); 698 // private static native void SR_VocabularySave(SR_Vocabulary* self, 699 // const LCHAR* filename); 700 // private static native void SR_VocabularyAddWord(SR_Vocabulary* self, 701 // const LCHAR* word); 702 // private static native void SR_VocabularyGetLanguage(SR_Vocabulary* self, 703 // ESR_Locale* locale); 704 private static native void SR_VocabularyDestroy(int vocabulary); 705 private static native String SR_VocabularyGetPronunciation(int vocabulary, String word); 706 707 708 // 709 // SR_RecognizerResult native methods 710 // 711 private static native byte[] SR_RecognizerResultGetWaveform(int recognizer); 712 private static native int SR_RecognizerResultGetSize(int recognizer); 713 private static native int SR_RecognizerResultGetKeyCount(int recognizer, int nbest); 714 private static native String[] SR_RecognizerResultGetKeyList(int recognizer, int nbest); 715 private static native String SR_RecognizerResultGetValue(int recognizer, 716 int nbest, String key); 717 // private static native void SR_RecognizerResultGetLocale(int recognizer, ESR_Locale* locale); 718 } 719
