TextToSpeech.java revision 952c0fd7bc6bb3ec9e0252f0799bbeabcf70dc1a
1/* 2 * Copyright (C) 2009 The Android Open Source Project 3 * 4 * Licensed under the Apache License, Version 2.0 (the "License"); you may not 5 * use this file except in compliance with the License. You may obtain a copy of 6 * the License at 7 * 8 * http://www.apache.org/licenses/LICENSE-2.0 9 * 10 * Unless required by applicable law or agreed to in writing, software 11 * distributed under the License is distributed on an "AS IS" BASIS, WITHOUT 12 * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the 13 * License for the specific language governing permissions and limitations under 14 * the License. 15 */ 16package android.speech.tts; 17 18import android.annotation.SdkConstant; 19import android.annotation.SdkConstant.SdkConstantType; 20import android.content.ComponentName; 21import android.content.ContentResolver; 22import android.content.Context; 23import android.content.Intent; 24import android.content.ServiceConnection; 25import android.media.AudioManager; 26import android.net.Uri; 27import android.os.AsyncTask; 28import android.os.Bundle; 29import android.os.IBinder; 30import android.os.ParcelFileDescriptor; 31import android.os.RemoteException; 32import android.provider.Settings; 33import android.text.TextUtils; 34import android.util.Log; 35 36import java.io.File; 37import java.io.FileNotFoundException; 38import java.io.IOException; 39import java.util.Collections; 40import java.util.HashMap; 41import java.util.HashSet; 42import java.util.List; 43import java.util.Locale; 44import java.util.Map; 45import java.util.MissingResourceException; 46import java.util.Set; 47 48/** 49 * 50 * Synthesizes speech from text for immediate playback or to create a sound file. 51 * <p>A TextToSpeech instance can only be used to synthesize text once it has completed its 52 * initialization. Implement the {@link TextToSpeech.OnInitListener} to be 53 * notified of the completion of the initialization.<br> 54 * When you are done using the TextToSpeech instance, call the {@link #shutdown()} method 55 * to release the native resources used by the TextToSpeech engine. 56 * 57 * @deprecated Use {@link TextToSpeechClient} instead 58 */ 59@Deprecated 60public class TextToSpeech { 61 62 private static final String TAG = "TextToSpeech"; 63 64 /** 65 * Denotes a successful operation. 66 */ 67 public static final int SUCCESS = 0; 68 /** 69 * Denotes a generic operation failure. 70 */ 71 public static final int ERROR = -1; 72 73 /** 74 * Queue mode where all entries in the playback queue (media to be played 75 * and text to be synthesized) are dropped and replaced by the new entry. 76 * Queues are flushed with respect to a given calling app. Entries in the queue 77 * from other callees are not discarded. 78 */ 79 public static final int QUEUE_FLUSH = 0; 80 /** 81 * Queue mode where the new entry is added at the end of the playback queue. 82 */ 83 public static final int QUEUE_ADD = 1; 84 /** 85 * Queue mode where the entire playback queue is purged. This is different 86 * from {@link #QUEUE_FLUSH} in that all entries are purged, not just entries 87 * from a given caller. 88 * 89 * @hide 90 */ 91 static final int QUEUE_DESTROY = 2; 92 93 /** 94 * Denotes the language is available exactly as specified by the locale. 95 */ 96 public static final int LANG_COUNTRY_VAR_AVAILABLE = 2; 97 98 /** 99 * Denotes the language is available for the language and country specified 100 * by the locale, but not the variant. 101 */ 102 public static final int LANG_COUNTRY_AVAILABLE = 1; 103 104 /** 105 * Denotes the language is available for the language by the locale, 106 * but not the country and variant. 107 */ 108 public static final int LANG_AVAILABLE = 0; 109 110 /** 111 * Denotes the language data is missing. 112 */ 113 public static final int LANG_MISSING_DATA = -1; 114 115 /** 116 * Denotes the language is not supported. 117 */ 118 public static final int LANG_NOT_SUPPORTED = -2; 119 120 /** 121 * Broadcast Action: The TextToSpeech synthesizer has completed processing 122 * of all the text in the speech queue. 123 * 124 * Note that this notifies callers when the <b>engine</b> has finished has 125 * processing text data. Audio playback might not have completed (or even started) 126 * at this point. If you wish to be notified when this happens, see 127 * {@link OnUtteranceCompletedListener}. 128 */ 129 @SdkConstant(SdkConstantType.BROADCAST_INTENT_ACTION) 130 public static final String ACTION_TTS_QUEUE_PROCESSING_COMPLETED = 131 "android.speech.tts.TTS_QUEUE_PROCESSING_COMPLETED"; 132 133 /** 134 * Interface definition of a callback to be invoked indicating the completion of the 135 * TextToSpeech engine initialization. 136 */ 137 public interface OnInitListener { 138 /** 139 * Called to signal the completion of the TextToSpeech engine initialization. 140 * 141 * @param status {@link TextToSpeech#SUCCESS} or {@link TextToSpeech#ERROR}. 142 */ 143 public void onInit(int status); 144 } 145 146 /** 147 * Listener that will be called when the TTS service has 148 * completed synthesizing an utterance. This is only called if the utterance 149 * has an utterance ID (see {@link TextToSpeech.Engine#KEY_PARAM_UTTERANCE_ID}). 150 * 151 * @deprecated Use {@link UtteranceProgressListener} instead. 152 */ 153 @Deprecated 154 public interface OnUtteranceCompletedListener { 155 /** 156 * Called when an utterance has been synthesized. 157 * 158 * @param utteranceId the identifier of the utterance. 159 */ 160 public void onUtteranceCompleted(String utteranceId); 161 } 162 163 /** 164 * Constants and parameter names for controlling text-to-speech. These include: 165 * 166 * <ul> 167 * <li> 168 * Intents to ask engine to install data or check its data and 169 * extras for a TTS engine's check data activity. 170 * </li> 171 * <li> 172 * Keys for the parameters passed with speak commands, e.g. 173 * {@link Engine#KEY_PARAM_UTTERANCE_ID}, {@link Engine#KEY_PARAM_STREAM}. 174 * </li> 175 * <li> 176 * A list of feature strings that engines might support, e.g 177 * {@link Engine#KEY_FEATURE_NETWORK_SYNTHESIS}). These values may be passed in to 178 * {@link TextToSpeech#speak} and {@link TextToSpeech#synthesizeToFile} to modify 179 * engine behaviour. The engine can be queried for the set of features it supports 180 * through {@link TextToSpeech#getFeatures(java.util.Locale)}. 181 * </li> 182 * </ul> 183 */ 184 public class Engine { 185 186 /** 187 * Default speech rate. 188 * @hide 189 */ 190 public static final int DEFAULT_RATE = 100; 191 192 /** 193 * Default pitch. 194 * @hide 195 */ 196 public static final int DEFAULT_PITCH = 100; 197 198 /** 199 * Default volume. 200 * @hide 201 */ 202 public static final float DEFAULT_VOLUME = 1.0f; 203 204 /** 205 * Default pan (centered). 206 * @hide 207 */ 208 public static final float DEFAULT_PAN = 0.0f; 209 210 /** 211 * Default value for {@link Settings.Secure#TTS_USE_DEFAULTS}. 212 * @hide 213 */ 214 public static final int USE_DEFAULTS = 0; // false 215 216 /** 217 * Package name of the default TTS engine. 218 * 219 * @hide 220 * @deprecated No longer in use, the default engine is determined by 221 * the sort order defined in {@link TtsEngines}. Note that 222 * this doesn't "break" anything because there is no guarantee that 223 * the engine specified below is installed on a given build, let 224 * alone be the default. 225 */ 226 @Deprecated 227 public static final String DEFAULT_ENGINE = "com.svox.pico"; 228 229 /** 230 * Default audio stream used when playing synthesized speech. 231 */ 232 public static final int DEFAULT_STREAM = AudioManager.STREAM_MUSIC; 233 234 /** 235 * Indicates success when checking the installation status of the resources used by the 236 * TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 237 */ 238 public static final int CHECK_VOICE_DATA_PASS = 1; 239 240 /** 241 * Indicates failure when checking the installation status of the resources used by the 242 * TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 243 */ 244 public static final int CHECK_VOICE_DATA_FAIL = 0; 245 246 /** 247 * Indicates erroneous data when checking the installation status of the resources used by 248 * the TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 249 * 250 * @deprecated Use CHECK_VOICE_DATA_FAIL instead. 251 */ 252 @Deprecated 253 public static final int CHECK_VOICE_DATA_BAD_DATA = -1; 254 255 /** 256 * Indicates missing resources when checking the installation status of the resources used 257 * by the TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 258 * 259 * @deprecated Use CHECK_VOICE_DATA_FAIL instead. 260 */ 261 @Deprecated 262 public static final int CHECK_VOICE_DATA_MISSING_DATA = -2; 263 264 /** 265 * Indicates missing storage volume when checking the installation status of the resources 266 * used by the TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 267 * 268 * @deprecated Use CHECK_VOICE_DATA_FAIL instead. 269 */ 270 @Deprecated 271 public static final int CHECK_VOICE_DATA_MISSING_VOLUME = -3; 272 273 /** 274 * Intent for starting a TTS service. Services that handle this intent must 275 * extend {@link TextToSpeechService}. Normal applications should not use this intent 276 * directly, instead they should talk to the TTS service using the the methods in this 277 * class. 278 */ 279 @SdkConstant(SdkConstantType.SERVICE_ACTION) 280 public static final String INTENT_ACTION_TTS_SERVICE = 281 "android.intent.action.TTS_SERVICE"; 282 283 /** 284 * Name under which a text to speech engine publishes information about itself. 285 * This meta-data should reference an XML resource containing a 286 * <code><{@link android.R.styleable#TextToSpeechEngine tts-engine}></code> 287 * tag. 288 */ 289 public static final String SERVICE_META_DATA = "android.speech.tts"; 290 291 // intents to ask engine to install data or check its data 292 /** 293 * Activity Action: Triggers the platform TextToSpeech engine to 294 * start the activity that installs the resource files on the device 295 * that are required for TTS to be operational. Since the installation 296 * of the data can be interrupted or declined by the user, the application 297 * shouldn't expect successful installation upon return from that intent, 298 * and if need be, should check installation status with 299 * {@link #ACTION_CHECK_TTS_DATA}. 300 */ 301 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 302 public static final String ACTION_INSTALL_TTS_DATA = 303 "android.speech.tts.engine.INSTALL_TTS_DATA"; 304 305 /** 306 * Broadcast Action: broadcast to signal the change in the list of available 307 * languages or/and their features. 308 */ 309 @SdkConstant(SdkConstantType.BROADCAST_INTENT_ACTION) 310 public static final String ACTION_TTS_DATA_INSTALLED = 311 "android.speech.tts.engine.TTS_DATA_INSTALLED"; 312 313 /** 314 * Activity Action: Starts the activity from the platform TextToSpeech 315 * engine to verify the proper installation and availability of the 316 * resource files on the system. Upon completion, the activity will 317 * return one of the following codes: 318 * {@link #CHECK_VOICE_DATA_PASS}, 319 * {@link #CHECK_VOICE_DATA_FAIL}, 320 * <p> Moreover, the data received in the activity result will contain the following 321 * fields: 322 * <ul> 323 * <li>{@link #EXTRA_AVAILABLE_VOICES} which contains an ArrayList<String> of all the 324 * available voices. The format of each voice is: lang-COUNTRY-variant where COUNTRY and 325 * variant are optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE").</li> 326 * <li>{@link #EXTRA_UNAVAILABLE_VOICES} which contains an ArrayList<String> of all the 327 * unavailable voices (ones that user can install). The format of each voice is: 328 * lang-COUNTRY-variant where COUNTRY and variant are optional (ie, "eng" or 329 * "eng-USA" or "eng-USA-FEMALE").</li> 330 * </ul> 331 */ 332 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 333 public static final String ACTION_CHECK_TTS_DATA = 334 "android.speech.tts.engine.CHECK_TTS_DATA"; 335 336 /** 337 * Activity intent for getting some sample text to use for demonstrating TTS. Specific 338 * locale have to be requested by passing following extra parameters: 339 * <ul> 340 * <li>language</li> 341 * <li>country</li> 342 * <li>variant</li> 343 * </ul> 344 * 345 * Upon completion, the activity result may contain the following fields: 346 * <ul> 347 * <li>{@link #EXTRA_SAMPLE_TEXT} which contains an String with sample text.</li> 348 * </ul> 349 */ 350 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 351 public static final String ACTION_GET_SAMPLE_TEXT = 352 "android.speech.tts.engine.GET_SAMPLE_TEXT"; 353 354 /** 355 * Extra information received with the {@link #ACTION_GET_SAMPLE_TEXT} intent result where 356 * the TextToSpeech engine returns an String with sample text for requested voice 357 */ 358 public static final String EXTRA_SAMPLE_TEXT = "sampleText"; 359 360 361 // extras for a TTS engine's check data activity 362 /** 363 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 364 * the TextToSpeech engine returns an ArrayList<String> of all the available voices. 365 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 366 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 367 */ 368 public static final String EXTRA_AVAILABLE_VOICES = "availableVoices"; 369 370 /** 371 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 372 * the TextToSpeech engine returns an ArrayList<String> of all the unavailable voices. 373 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 374 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 375 */ 376 public static final String EXTRA_UNAVAILABLE_VOICES = "unavailableVoices"; 377 378 /** 379 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 380 * the TextToSpeech engine specifies the path to its resources. 381 * 382 * It may be used by language packages to find out where to put their data. 383 * 384 * @deprecated TTS engine implementation detail, this information has no use for 385 * text-to-speech API client. 386 */ 387 @Deprecated 388 public static final String EXTRA_VOICE_DATA_ROOT_DIRECTORY = "dataRoot"; 389 390 /** 391 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 392 * the TextToSpeech engine specifies the file names of its resources under the 393 * resource path. 394 * 395 * @deprecated TTS engine implementation detail, this information has no use for 396 * text-to-speech API client. 397 */ 398 @Deprecated 399 public static final String EXTRA_VOICE_DATA_FILES = "dataFiles"; 400 401 /** 402 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 403 * the TextToSpeech engine specifies the locale associated with each resource file. 404 * 405 * @deprecated TTS engine implementation detail, this information has no use for 406 * text-to-speech API client. 407 */ 408 @Deprecated 409 public static final String EXTRA_VOICE_DATA_FILES_INFO = "dataFilesInfo"; 410 411 /** 412 * Extra information sent with the {@link #ACTION_CHECK_TTS_DATA} intent where the 413 * caller indicates to the TextToSpeech engine which specific sets of voice data to 414 * check for by sending an ArrayList<String> of the voices that are of interest. 415 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 416 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 417 * 418 * @deprecated Redundant functionality, checking for existence of specific sets of voice 419 * data can be done on client side. 420 */ 421 @Deprecated 422 public static final String EXTRA_CHECK_VOICE_DATA_FOR = "checkVoiceDataFor"; 423 424 // extras for a TTS engine's data installation 425 /** 426 * Extra information received with the {@link #ACTION_TTS_DATA_INSTALLED} intent result. 427 * It indicates whether the data files for the synthesis engine were successfully 428 * installed. The installation was initiated with the {@link #ACTION_INSTALL_TTS_DATA} 429 * intent. The possible values for this extra are 430 * {@link TextToSpeech#SUCCESS} and {@link TextToSpeech#ERROR}. 431 * 432 * @deprecated No longer in use. If client ise interested in information about what 433 * changed, is should send ACTION_CHECK_TTS_DATA intent to discover available voices. 434 */ 435 @Deprecated 436 public static final String EXTRA_TTS_DATA_INSTALLED = "dataInstalled"; 437 438 // keys for the parameters passed with speak commands. Hidden keys are used internally 439 // to maintain engine state for each TextToSpeech instance. 440 /** 441 * @hide 442 */ 443 public static final String KEY_PARAM_RATE = "rate"; 444 445 /** 446 * @hide 447 */ 448 public static final String KEY_PARAM_LANGUAGE = "language"; 449 450 /** 451 * @hide 452 */ 453 public static final String KEY_PARAM_COUNTRY = "country"; 454 455 /** 456 * @hide 457 */ 458 public static final String KEY_PARAM_VARIANT = "variant"; 459 460 /** 461 * @hide 462 */ 463 public static final String KEY_PARAM_ENGINE = "engine"; 464 465 /** 466 * @hide 467 */ 468 public static final String KEY_PARAM_PITCH = "pitch"; 469 470 /** 471 * Parameter key to specify the audio stream type to be used when speaking text 472 * or playing back a file. The value should be one of the STREAM_ constants 473 * defined in {@link AudioManager}. 474 * 475 * @see TextToSpeech#speak(String, int, HashMap) 476 * @see TextToSpeech#playEarcon(String, int, HashMap) 477 */ 478 public static final String KEY_PARAM_STREAM = "streamType"; 479 480 /** 481 * Parameter key to identify an utterance in the 482 * {@link TextToSpeech.OnUtteranceCompletedListener} after text has been 483 * spoken, a file has been played back or a silence duration has elapsed. 484 * 485 * @see TextToSpeech#speak(String, int, HashMap) 486 * @see TextToSpeech#playEarcon(String, int, HashMap) 487 * @see TextToSpeech#synthesizeToFile(String, HashMap, String) 488 */ 489 public static final String KEY_PARAM_UTTERANCE_ID = "utteranceId"; 490 491 /** 492 * Parameter key to specify the speech volume relative to the current stream type 493 * volume used when speaking text. Volume is specified as a float ranging from 0 to 1 494 * where 0 is silence, and 1 is the maximum volume (the default behavior). 495 * 496 * @see TextToSpeech#speak(String, int, HashMap) 497 * @see TextToSpeech#playEarcon(String, int, HashMap) 498 */ 499 public static final String KEY_PARAM_VOLUME = "volume"; 500 501 /** 502 * Parameter key to specify how the speech is panned from left to right when speaking text. 503 * Pan is specified as a float ranging from -1 to +1 where -1 maps to a hard-left pan, 504 * 0 to center (the default behavior), and +1 to hard-right. 505 * 506 * @see TextToSpeech#speak(String, int, HashMap) 507 * @see TextToSpeech#playEarcon(String, int, HashMap) 508 */ 509 public static final String KEY_PARAM_PAN = "pan"; 510 511 /** 512 * Feature key for network synthesis. See {@link TextToSpeech#getFeatures(Locale)} 513 * for a description of how feature keys work. If set (and supported by the engine 514 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must 515 * use network based synthesis. 516 * 517 * @see TextToSpeech#speak(String, int, java.util.HashMap) 518 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 519 * @see TextToSpeech#getFeatures(java.util.Locale) 520 */ 521 public static final String KEY_FEATURE_NETWORK_SYNTHESIS = "networkTts"; 522 523 /** 524 * Feature key for embedded synthesis. See {@link TextToSpeech#getFeatures(Locale)} 525 * for a description of how feature keys work. If set and supported by the engine 526 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must synthesize 527 * text on-device (without making network requests). 528 * 529 * @see TextToSpeech#speak(String, int, java.util.HashMap) 530 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 531 * @see TextToSpeech#getFeatures(java.util.Locale) 532 */ 533 public static final String KEY_FEATURE_EMBEDDED_SYNTHESIS = "embeddedTts"; 534 } 535 536 private final Context mContext; 537 private Connection mConnectingServiceConnection; 538 private Connection mServiceConnection; 539 private OnInitListener mInitListener; 540 // Written from an unspecified application thread, read from 541 // a binder thread. 542 private volatile UtteranceProgressListener mUtteranceProgressListener; 543 private final Object mStartLock = new Object(); 544 545 private String mRequestedEngine; 546 // Whether to initialize this TTS object with the default engine, 547 // if the requested engine is not available. Valid only if mRequestedEngine 548 // is not null. Used only for testing, though potentially useful API wise 549 // too. 550 private final boolean mUseFallback; 551 private final Map<String, Uri> mEarcons; 552 private final Map<String, Uri> mUtterances; 553 private final Bundle mParams = new Bundle(); 554 private final TtsEngines mEnginesHelper; 555 private final String mPackageName; 556 private volatile String mCurrentEngine = null; 557 558 /** 559 * The constructor for the TextToSpeech class, using the default TTS engine. 560 * This will also initialize the associated TextToSpeech engine if it isn't already running. 561 * 562 * @param context 563 * The context this instance is running in. 564 * @param listener 565 * The {@link TextToSpeech.OnInitListener} that will be called when the 566 * TextToSpeech engine has initialized. In a case of a failure the listener 567 * may be called immediately, before TextToSpeech instance is fully constructed. 568 */ 569 public TextToSpeech(Context context, OnInitListener listener) { 570 this(context, listener, null); 571 } 572 573 /** 574 * The constructor for the TextToSpeech class, using the given TTS engine. 575 * This will also initialize the associated TextToSpeech engine if it isn't already running. 576 * 577 * @param context 578 * The context this instance is running in. 579 * @param listener 580 * The {@link TextToSpeech.OnInitListener} that will be called when the 581 * TextToSpeech engine has initialized. In a case of a failure the listener 582 * may be called immediately, before TextToSpeech instance is fully constructed. 583 * @param engine Package name of the TTS engine to use. 584 */ 585 public TextToSpeech(Context context, OnInitListener listener, String engine) { 586 this(context, listener, engine, null, true); 587 } 588 589 /** 590 * Used by the framework to instantiate TextToSpeech objects with a supplied 591 * package name, instead of using {@link android.content.Context#getPackageName()} 592 * 593 * @hide 594 */ 595 public TextToSpeech(Context context, OnInitListener listener, String engine, 596 String packageName, boolean useFallback) { 597 mContext = context; 598 mInitListener = listener; 599 mRequestedEngine = engine; 600 mUseFallback = useFallback; 601 602 mEarcons = new HashMap<String, Uri>(); 603 mUtterances = new HashMap<String, Uri>(); 604 mUtteranceProgressListener = null; 605 606 mEnginesHelper = new TtsEngines(mContext); 607 if (packageName != null) { 608 mPackageName = packageName; 609 } else { 610 mPackageName = mContext.getPackageName(); 611 } 612 initTts(); 613 } 614 615 private <R> R runActionNoReconnect(Action<R> action, R errorResult, String method, 616 boolean onlyEstablishedConnection) { 617 return runAction(action, errorResult, method, false, onlyEstablishedConnection); 618 } 619 620 private <R> R runAction(Action<R> action, R errorResult, String method) { 621 return runAction(action, errorResult, method, true, true); 622 } 623 624 private <R> R runAction(Action<R> action, R errorResult, String method, 625 boolean reconnect, boolean onlyEstablishedConnection) { 626 synchronized (mStartLock) { 627 if (mServiceConnection == null) { 628 Log.w(TAG, method + " failed: not bound to TTS engine"); 629 return errorResult; 630 } 631 return mServiceConnection.runAction(action, errorResult, method, reconnect, 632 onlyEstablishedConnection); 633 } 634 } 635 636 private int initTts() { 637 // Step 1: Try connecting to the engine that was requested. 638 if (mRequestedEngine != null) { 639 if (mEnginesHelper.isEngineInstalled(mRequestedEngine)) { 640 if (connectToEngine(mRequestedEngine)) { 641 mCurrentEngine = mRequestedEngine; 642 return SUCCESS; 643 } else if (!mUseFallback) { 644 mCurrentEngine = null; 645 dispatchOnInit(ERROR); 646 return ERROR; 647 } 648 } else if (!mUseFallback) { 649 Log.i(TAG, "Requested engine not installed: " + mRequestedEngine); 650 mCurrentEngine = null; 651 dispatchOnInit(ERROR); 652 return ERROR; 653 } 654 } 655 656 // Step 2: Try connecting to the user's default engine. 657 final String defaultEngine = getDefaultEngine(); 658 if (defaultEngine != null && !defaultEngine.equals(mRequestedEngine)) { 659 if (connectToEngine(defaultEngine)) { 660 mCurrentEngine = defaultEngine; 661 return SUCCESS; 662 } 663 } 664 665 // Step 3: Try connecting to the highest ranked engine in the 666 // system. 667 final String highestRanked = mEnginesHelper.getHighestRankedEngineName(); 668 if (highestRanked != null && !highestRanked.equals(mRequestedEngine) && 669 !highestRanked.equals(defaultEngine)) { 670 if (connectToEngine(highestRanked)) { 671 mCurrentEngine = highestRanked; 672 return SUCCESS; 673 } 674 } 675 676 // NOTE: The API currently does not allow the caller to query whether 677 // they are actually connected to any engine. This might fail for various 678 // reasons like if the user disables all her TTS engines. 679 680 mCurrentEngine = null; 681 dispatchOnInit(ERROR); 682 return ERROR; 683 } 684 685 private boolean connectToEngine(String engine) { 686 Connection connection = new Connection(); 687 Intent intent = new Intent(Engine.INTENT_ACTION_TTS_SERVICE); 688 intent.setPackage(engine); 689 boolean bound = mContext.bindService(intent, connection, Context.BIND_AUTO_CREATE); 690 if (!bound) { 691 Log.e(TAG, "Failed to bind to " + engine); 692 return false; 693 } else { 694 Log.i(TAG, "Sucessfully bound to " + engine); 695 mConnectingServiceConnection = connection; 696 return true; 697 } 698 } 699 700 private void dispatchOnInit(int result) { 701 synchronized (mStartLock) { 702 if (mInitListener != null) { 703 mInitListener.onInit(result); 704 mInitListener = null; 705 } 706 } 707 } 708 709 private IBinder getCallerIdentity() { 710 return mServiceConnection.getCallerIdentity(); 711 } 712 713 /** 714 * Releases the resources used by the TextToSpeech engine. 715 * It is good practice for instance to call this method in the onDestroy() method of an Activity 716 * so the TextToSpeech engine can be cleanly stopped. 717 */ 718 public void shutdown() { 719 // Special case, we are asked to shutdown connection that did finalize its connection. 720 synchronized (mStartLock) { 721 if (mConnectingServiceConnection != null) { 722 mContext.unbindService(mConnectingServiceConnection); 723 mConnectingServiceConnection = null; 724 return; 725 } 726 } 727 728 // Post connection case 729 runActionNoReconnect(new Action<Void>() { 730 @Override 731 public Void run(ITextToSpeechService service) throws RemoteException { 732 service.setCallback(getCallerIdentity(), null); 733 service.stop(getCallerIdentity()); 734 mServiceConnection.disconnect(); 735 // Context#unbindService does not result in a call to 736 // ServiceConnection#onServiceDisconnected. As a result, the 737 // service ends up being destroyed (if there are no other open 738 // connections to it) but the process lives on and the 739 // ServiceConnection continues to refer to the destroyed service. 740 // 741 // This leads to tons of log spam about SynthThread being dead. 742 mServiceConnection = null; 743 mCurrentEngine = null; 744 return null; 745 } 746 }, null, "shutdown", false); 747 } 748 749 /** 750 * Adds a mapping between a string of text and a sound resource in a 751 * package. After a call to this method, subsequent calls to 752 * {@link #speak(String, int, HashMap)} will play the specified sound resource 753 * if it is available, or synthesize the text it is missing. 754 * 755 * @param text 756 * The string of text. Example: <code>"south_south_east"</code> 757 * 758 * @param packagename 759 * Pass the packagename of the application that contains the 760 * resource. If the resource is in your own application (this is 761 * the most common case), then put the packagename of your 762 * application here.<br/> 763 * Example: <b>"com.google.marvin.compass"</b><br/> 764 * The packagename can be found in the AndroidManifest.xml of 765 * your application. 766 * <p> 767 * <code><manifest xmlns:android="..." 768 * package="<b>com.google.marvin.compass</b>"></code> 769 * </p> 770 * 771 * @param resourceId 772 * Example: <code>R.raw.south_south_east</code> 773 * 774 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 775 */ 776 public int addSpeech(String text, String packagename, int resourceId) { 777 synchronized (mStartLock) { 778 mUtterances.put(text, makeResourceUri(packagename, resourceId)); 779 return SUCCESS; 780 } 781 } 782 783 /** 784 * Adds a mapping between a string of text and a sound file. Using this, it 785 * is possible to add custom pronounciations for a string of text. 786 * After a call to this method, subsequent calls to {@link #speak(String, int, HashMap)} 787 * will play the specified sound resource if it is available, or synthesize the text it is 788 * missing. 789 * 790 * @param text 791 * The string of text. Example: <code>"south_south_east"</code> 792 * @param filename 793 * The full path to the sound file (for example: 794 * "/sdcard/mysounds/hello.wav") 795 * 796 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 797 */ 798 public int addSpeech(String text, String filename) { 799 synchronized (mStartLock) { 800 mUtterances.put(text, Uri.parse(filename)); 801 return SUCCESS; 802 } 803 } 804 805 806 /** 807 * Adds a mapping between a string of text and a sound resource in a 808 * package. Use this to add custom earcons. 809 * 810 * @see #playEarcon(String, int, HashMap) 811 * 812 * @param earcon The name of the earcon. 813 * Example: <code>"[tick]"</code><br/> 814 * 815 * @param packagename 816 * the package name of the application that contains the 817 * resource. This can for instance be the package name of your own application. 818 * Example: <b>"com.google.marvin.compass"</b><br/> 819 * The package name can be found in the AndroidManifest.xml of 820 * the application containing the resource. 821 * <p> 822 * <code><manifest xmlns:android="..." 823 * package="<b>com.google.marvin.compass</b>"></code> 824 * </p> 825 * 826 * @param resourceId 827 * Example: <code>R.raw.tick_snd</code> 828 * 829 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 830 */ 831 public int addEarcon(String earcon, String packagename, int resourceId) { 832 synchronized(mStartLock) { 833 mEarcons.put(earcon, makeResourceUri(packagename, resourceId)); 834 return SUCCESS; 835 } 836 } 837 838 /** 839 * Adds a mapping between a string of text and a sound file. 840 * Use this to add custom earcons. 841 * 842 * @see #playEarcon(String, int, HashMap) 843 * 844 * @param earcon 845 * The name of the earcon. 846 * Example: <code>"[tick]"</code> 847 * @param filename 848 * The full path to the sound file (for example: 849 * "/sdcard/mysounds/tick.wav") 850 * 851 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 852 */ 853 public int addEarcon(String earcon, String filename) { 854 synchronized(mStartLock) { 855 mEarcons.put(earcon, Uri.parse(filename)); 856 return SUCCESS; 857 } 858 } 859 860 private Uri makeResourceUri(String packageName, int resourceId) { 861 return new Uri.Builder() 862 .scheme(ContentResolver.SCHEME_ANDROID_RESOURCE) 863 .encodedAuthority(packageName) 864 .appendEncodedPath(String.valueOf(resourceId)) 865 .build(); 866 } 867 868 /** 869 * Speaks the string using the specified queuing strategy and speech parameters. 870 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 871 * requests and then returns. The synthesis might not have finished (or even started!) at the 872 * time when this method returns. In order to reliably detect errors during synthesis, 873 * we recommend setting an utterance progress listener (see 874 * {@link #setOnUtteranceProgressListener}) and using the 875 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 876 * 877 * @param text The string of text to be spoken. No longer than 878 * {@link #getMaxSpeechInputLength()} characters. 879 * @param queueMode The queuing strategy to use, {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 880 * @param params Parameters for the request. Can be null. 881 * Supported parameter names: 882 * {@link Engine#KEY_PARAM_STREAM}, 883 * {@link Engine#KEY_PARAM_UTTERANCE_ID}, 884 * {@link Engine#KEY_PARAM_VOLUME}, 885 * {@link Engine#KEY_PARAM_PAN}. 886 * Engine specific parameters may be passed in but the parameter keys 887 * must be prefixed by the name of the engine they are intended for. For example 888 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 889 * engine named "com.svox.pico" if it is being used. 890 * 891 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the speak operation. 892 */ 893 public int speak(final String text, final int queueMode, final HashMap<String, String> params) { 894 return runAction(new Action<Integer>() { 895 @Override 896 public Integer run(ITextToSpeechService service) throws RemoteException { 897 Uri utteranceUri = mUtterances.get(text); 898 if (utteranceUri != null) { 899 return service.playAudio(getCallerIdentity(), utteranceUri, queueMode, 900 getParams(params)); 901 } else { 902 return service.speak(getCallerIdentity(), text, queueMode, getParams(params)); 903 } 904 } 905 }, ERROR, "speak"); 906 } 907 908 /** 909 * Plays the earcon using the specified queueing mode and parameters. 910 * The earcon must already have been added with {@link #addEarcon(String, String)} or 911 * {@link #addEarcon(String, String, int)}. 912 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 913 * requests and then returns. The synthesis might not have finished (or even started!) at the 914 * time when this method returns. In order to reliably detect errors during synthesis, 915 * we recommend setting an utterance progress listener (see 916 * {@link #setOnUtteranceProgressListener}) and using the 917 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 918 * 919 * @param earcon The earcon that should be played 920 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 921 * @param params Parameters for the request. Can be null. 922 * Supported parameter names: 923 * {@link Engine#KEY_PARAM_STREAM}, 924 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 925 * Engine specific parameters may be passed in but the parameter keys 926 * must be prefixed by the name of the engine they are intended for. For example 927 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 928 * engine named "com.svox.pico" if it is being used. 929 * 930 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playEarcon operation. 931 */ 932 public int playEarcon(final String earcon, final int queueMode, 933 final HashMap<String, String> params) { 934 return runAction(new Action<Integer>() { 935 @Override 936 public Integer run(ITextToSpeechService service) throws RemoteException { 937 Uri earconUri = mEarcons.get(earcon); 938 if (earconUri == null) { 939 return ERROR; 940 } 941 return service.playAudio(getCallerIdentity(), earconUri, queueMode, 942 getParams(params)); 943 } 944 }, ERROR, "playEarcon"); 945 } 946 947 /** 948 * Plays silence for the specified amount of time using the specified 949 * queue mode. 950 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 951 * requests and then returns. The synthesis might not have finished (or even started!) at the 952 * time when this method returns. In order to reliably detect errors during synthesis, 953 * we recommend setting an utterance progress listener (see 954 * {@link #setOnUtteranceProgressListener}) and using the 955 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 956 * 957 * @param durationInMs The duration of the silence. 958 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 959 * @param params Parameters for the request. Can be null. 960 * Supported parameter names: 961 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 962 * Engine specific parameters may be passed in but the parameter keys 963 * must be prefixed by the name of the engine they are intended for. For example 964 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 965 * engine named "com.svox.pico" if it is being used. 966 * 967 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playSilence operation. 968 */ 969 public int playSilence(final long durationInMs, final int queueMode, 970 final HashMap<String, String> params) { 971 return runAction(new Action<Integer>() { 972 @Override 973 public Integer run(ITextToSpeechService service) throws RemoteException { 974 return service.playSilence(getCallerIdentity(), durationInMs, queueMode, 975 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 976 } 977 }, ERROR, "playSilence"); 978 } 979 980 /** 981 * Queries the engine for the set of features it supports for a given locale. 982 * Features can either be framework defined, e.g. 983 * {@link TextToSpeech.Engine#KEY_FEATURE_NETWORK_SYNTHESIS} or engine specific. 984 * Engine specific keys must be prefixed by the name of the engine they 985 * are intended for. These keys can be used as parameters to 986 * {@link TextToSpeech#speak(String, int, java.util.HashMap)} and 987 * {@link TextToSpeech#synthesizeToFile(String, java.util.HashMap, String)}. 988 * 989 * Features are boolean flags, and their values in the synthesis parameters 990 * must be behave as per {@link Boolean#parseBoolean(String)}. 991 * 992 * @param locale The locale to query features for. 993 */ 994 public Set<String> getFeatures(final Locale locale) { 995 return runAction(new Action<Set<String>>() { 996 @Override 997 public Set<String> run(ITextToSpeechService service) throws RemoteException { 998 String[] features = null; 999 try { 1000 features = service.getFeaturesForLanguage( 1001 locale.getISO3Language(), locale.getISO3Country(), locale.getVariant()); 1002 } catch(MissingResourceException e) { 1003 Log.w(TAG, "Couldn't retrieve 3 letter ISO 639-2/T language and/or ISO 3166 " + 1004 "country code for locale: " + locale, e); 1005 return null; 1006 } 1007 1008 if (features != null) { 1009 final Set<String> featureSet = new HashSet<String>(); 1010 Collections.addAll(featureSet, features); 1011 return featureSet; 1012 } 1013 return null; 1014 } 1015 }, null, "getFeatures"); 1016 } 1017 1018 /** 1019 * Checks whether the TTS engine is busy speaking. Note that a speech item is 1020 * considered complete once it's audio data has been sent to the audio mixer, or 1021 * written to a file. There might be a finite lag between this point, and when 1022 * the audio hardware completes playback. 1023 * 1024 * @return {@code true} if the TTS engine is speaking. 1025 */ 1026 public boolean isSpeaking() { 1027 return runAction(new Action<Boolean>() { 1028 @Override 1029 public Boolean run(ITextToSpeechService service) throws RemoteException { 1030 return service.isSpeaking(); 1031 } 1032 }, false, "isSpeaking"); 1033 } 1034 1035 /** 1036 * Interrupts the current utterance (whether played or rendered to file) and discards other 1037 * utterances in the queue. 1038 * 1039 * @return {@link #ERROR} or {@link #SUCCESS}. 1040 */ 1041 public int stop() { 1042 return runAction(new Action<Integer>() { 1043 @Override 1044 public Integer run(ITextToSpeechService service) throws RemoteException { 1045 return service.stop(getCallerIdentity()); 1046 } 1047 }, ERROR, "stop"); 1048 } 1049 1050 /** 1051 * Sets the speech rate. 1052 * 1053 * This has no effect on any pre-recorded speech. 1054 * 1055 * @param speechRate Speech rate. {@code 1.0} is the normal speech rate, 1056 * lower values slow down the speech ({@code 0.5} is half the normal speech rate), 1057 * greater values accelerate it ({@code 2.0} is twice the normal speech rate). 1058 * 1059 * @return {@link #ERROR} or {@link #SUCCESS}. 1060 */ 1061 public int setSpeechRate(float speechRate) { 1062 if (speechRate > 0.0f) { 1063 int intRate = (int)(speechRate * 100); 1064 if (intRate > 0) { 1065 synchronized (mStartLock) { 1066 mParams.putInt(Engine.KEY_PARAM_RATE, intRate); 1067 } 1068 return SUCCESS; 1069 } 1070 } 1071 return ERROR; 1072 } 1073 1074 /** 1075 * Sets the speech pitch for the TextToSpeech engine. 1076 * 1077 * This has no effect on any pre-recorded speech. 1078 * 1079 * @param pitch Speech pitch. {@code 1.0} is the normal pitch, 1080 * lower values lower the tone of the synthesized voice, 1081 * greater values increase it. 1082 * 1083 * @return {@link #ERROR} or {@link #SUCCESS}. 1084 */ 1085 public int setPitch(float pitch) { 1086 if (pitch > 0.0f) { 1087 int intPitch = (int)(pitch * 100); 1088 if (intPitch > 0) { 1089 synchronized (mStartLock) { 1090 mParams.putInt(Engine.KEY_PARAM_PITCH, intPitch); 1091 } 1092 return SUCCESS; 1093 } 1094 } 1095 return ERROR; 1096 } 1097 1098 /** 1099 * @return the engine currently in use by this TextToSpeech instance. 1100 * @hide 1101 */ 1102 public String getCurrentEngine() { 1103 return mCurrentEngine; 1104 } 1105 1106 /** 1107 * Returns a Locale instance describing the language currently being used as the default 1108 * Text-to-speech language. 1109 * 1110 * @return language, country (if any) and variant (if any) used by the client stored in a 1111 * Locale instance, or {@code null} on error. 1112 */ 1113 public Locale getDefaultLanguage() { 1114 return runAction(new Action<Locale>() { 1115 @Override 1116 public Locale run(ITextToSpeechService service) throws RemoteException { 1117 String[] defaultLanguage = service.getClientDefaultLanguage(); 1118 1119 return new Locale(defaultLanguage[0], defaultLanguage[1], defaultLanguage[2]); 1120 } 1121 }, null, "getDefaultLanguage"); 1122 } 1123 1124 /** 1125 * Sets the text-to-speech language. 1126 * The TTS engine will try to use the closest match to the specified 1127 * language as represented by the Locale, but there is no guarantee that the exact same Locale 1128 * will be used. Use {@link #isLanguageAvailable(Locale)} to check the level of support 1129 * before choosing the language to use for the next utterances. 1130 * 1131 * @param loc The locale describing the language to be used. 1132 * 1133 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1134 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1135 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1136 */ 1137 public int setLanguage(final Locale loc) { 1138 return runAction(new Action<Integer>() { 1139 @Override 1140 public Integer run(ITextToSpeechService service) throws RemoteException { 1141 if (loc == null) { 1142 return LANG_NOT_SUPPORTED; 1143 } 1144 String language = null, country = null; 1145 try { 1146 language = loc.getISO3Language(); 1147 } catch (MissingResourceException e) { 1148 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1149 return LANG_NOT_SUPPORTED; 1150 } 1151 1152 try { 1153 country = loc.getISO3Country(); 1154 } catch (MissingResourceException e) { 1155 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1156 return LANG_NOT_SUPPORTED; 1157 } 1158 1159 String variant = loc.getVariant(); 1160 1161 // Check if the language, country, variant are available, and cache 1162 // the available parts. 1163 // Note that the language is not actually set here, instead it is cached so it 1164 // will be associated with all upcoming utterances. 1165 1166 int result = service.loadLanguage(getCallerIdentity(), language, country, variant); 1167 if (result >= LANG_AVAILABLE){ 1168 if (result < LANG_COUNTRY_VAR_AVAILABLE) { 1169 variant = ""; 1170 if (result < LANG_COUNTRY_AVAILABLE) { 1171 country = ""; 1172 } 1173 } 1174 mParams.putString(Engine.KEY_PARAM_LANGUAGE, language); 1175 mParams.putString(Engine.KEY_PARAM_COUNTRY, country); 1176 mParams.putString(Engine.KEY_PARAM_VARIANT, variant); 1177 } 1178 return result; 1179 } 1180 }, LANG_NOT_SUPPORTED, "setLanguage"); 1181 } 1182 1183 /** 1184 * Returns a Locale instance describing the language currently being used for synthesis 1185 * requests sent to the TextToSpeech engine. 1186 * 1187 * In Android 4.2 and before (API <= 17) this function returns the language that is currently 1188 * being used by the TTS engine. That is the last language set by this or any other 1189 * client by a {@link TextToSpeech#setLanguage} call to the same engine. 1190 * 1191 * In Android versions after 4.2 this function returns the language that is currently being 1192 * used for the synthesis requests sent from this client. That is the last language set 1193 * by a {@link TextToSpeech#setLanguage} call on this instance. 1194 * 1195 * @return language, country (if any) and variant (if any) used by the client stored in a 1196 * Locale instance, or {@code null} on error. 1197 */ 1198 public Locale getLanguage() { 1199 return runAction(new Action<Locale>() { 1200 @Override 1201 public Locale run(ITextToSpeechService service) { 1202 /* No service call, but we're accessing mParams, hence need for 1203 wrapping it as an Action instance */ 1204 String lang = mParams.getString(Engine.KEY_PARAM_LANGUAGE, ""); 1205 String country = mParams.getString(Engine.KEY_PARAM_COUNTRY, ""); 1206 String variant = mParams.getString(Engine.KEY_PARAM_VARIANT, ""); 1207 return new Locale(lang, country, variant); 1208 } 1209 }, null, "getLanguage"); 1210 } 1211 1212 /** 1213 * Checks if the specified language as represented by the Locale is available and supported. 1214 * 1215 * @param loc The Locale describing the language to be used. 1216 * 1217 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1218 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1219 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1220 */ 1221 public int isLanguageAvailable(final Locale loc) { 1222 return runAction(new Action<Integer>() { 1223 @Override 1224 public Integer run(ITextToSpeechService service) throws RemoteException { 1225 String language = null, country = null; 1226 1227 try { 1228 language = loc.getISO3Language(); 1229 } catch (MissingResourceException e) { 1230 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1231 return LANG_NOT_SUPPORTED; 1232 } 1233 1234 try { 1235 country = loc.getISO3Country(); 1236 } catch (MissingResourceException e) { 1237 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1238 return LANG_NOT_SUPPORTED; 1239 } 1240 1241 return service.isLanguageAvailable(language, country, loc.getVariant()); 1242 } 1243 }, LANG_NOT_SUPPORTED, "isLanguageAvailable"); 1244 } 1245 1246 /** 1247 * Synthesizes the given text to a file using the specified parameters. 1248 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1249 * requests and then returns. The synthesis might not have finished (or even started!) at the 1250 * time when this method returns. In order to reliably detect errors during synthesis, 1251 * we recommend setting an utterance progress listener (see 1252 * {@link #setOnUtteranceProgressListener}) and using the 1253 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1254 * 1255 * @param text The text that should be synthesized. No longer than 1256 * {@link #getMaxSpeechInputLength()} characters. 1257 * @param params Parameters for the request. Can be null. 1258 * Supported parameter names: 1259 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1260 * Engine specific parameters may be passed in but the parameter keys 1261 * must be prefixed by the name of the engine they are intended for. For example 1262 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1263 * engine named "com.svox.pico" if it is being used. 1264 * @param filename Absolute file filename to write the generated audio data to.It should be 1265 * something like "/sdcard/myappsounds/mysound.wav". 1266 * 1267 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the synthesizeToFile operation. 1268 */ 1269 public int synthesizeToFile(final String text, final HashMap<String, String> params, 1270 final String filename) { 1271 return runAction(new Action<Integer>() { 1272 @Override 1273 public Integer run(ITextToSpeechService service) throws RemoteException { 1274 ParcelFileDescriptor fileDescriptor; 1275 int returnValue; 1276 try { 1277 File file = new File(filename); 1278 if(file.exists() && !file.canWrite()) { 1279 Log.e(TAG, "Can't write to " + filename); 1280 return ERROR; 1281 } 1282 fileDescriptor = ParcelFileDescriptor.open(file, 1283 ParcelFileDescriptor.MODE_WRITE_ONLY | 1284 ParcelFileDescriptor.MODE_CREATE | 1285 ParcelFileDescriptor.MODE_TRUNCATE); 1286 returnValue = service.synthesizeToFileDescriptor(getCallerIdentity(), text, 1287 fileDescriptor, getParams(params)); 1288 fileDescriptor.close(); 1289 return returnValue; 1290 } catch (FileNotFoundException e) { 1291 Log.e(TAG, "Opening file " + filename + " failed", e); 1292 return ERROR; 1293 } catch (IOException e) { 1294 Log.e(TAG, "Closing file " + filename + " failed", e); 1295 return ERROR; 1296 } 1297 } 1298 }, ERROR, "synthesizeToFile"); 1299 } 1300 1301 private Bundle getParams(HashMap<String, String> params) { 1302 if (params != null && !params.isEmpty()) { 1303 Bundle bundle = new Bundle(mParams); 1304 copyIntParam(bundle, params, Engine.KEY_PARAM_STREAM); 1305 copyStringParam(bundle, params, Engine.KEY_PARAM_UTTERANCE_ID); 1306 copyFloatParam(bundle, params, Engine.KEY_PARAM_VOLUME); 1307 copyFloatParam(bundle, params, Engine.KEY_PARAM_PAN); 1308 1309 // Copy feature strings defined by the framework. 1310 copyStringParam(bundle, params, Engine.KEY_FEATURE_NETWORK_SYNTHESIS); 1311 copyStringParam(bundle, params, Engine.KEY_FEATURE_EMBEDDED_SYNTHESIS); 1312 1313 // Copy over all parameters that start with the name of the 1314 // engine that we are currently connected to. The engine is 1315 // free to interpret them as it chooses. 1316 if (!TextUtils.isEmpty(mCurrentEngine)) { 1317 for (Map.Entry<String, String> entry : params.entrySet()) { 1318 final String key = entry.getKey(); 1319 if (key != null && key.startsWith(mCurrentEngine)) { 1320 bundle.putString(key, entry.getValue()); 1321 } 1322 } 1323 } 1324 1325 return bundle; 1326 } else { 1327 return mParams; 1328 } 1329 } 1330 1331 private void copyStringParam(Bundle bundle, HashMap<String, String> params, String key) { 1332 String value = params.get(key); 1333 if (value != null) { 1334 bundle.putString(key, value); 1335 } 1336 } 1337 1338 private void copyIntParam(Bundle bundle, HashMap<String, String> params, String key) { 1339 String valueString = params.get(key); 1340 if (!TextUtils.isEmpty(valueString)) { 1341 try { 1342 int value = Integer.parseInt(valueString); 1343 bundle.putInt(key, value); 1344 } catch (NumberFormatException ex) { 1345 // don't set the value in the bundle 1346 } 1347 } 1348 } 1349 1350 private void copyFloatParam(Bundle bundle, HashMap<String, String> params, String key) { 1351 String valueString = params.get(key); 1352 if (!TextUtils.isEmpty(valueString)) { 1353 try { 1354 float value = Float.parseFloat(valueString); 1355 bundle.putFloat(key, value); 1356 } catch (NumberFormatException ex) { 1357 // don't set the value in the bundle 1358 } 1359 } 1360 } 1361 1362 /** 1363 * Sets the listener that will be notified when synthesis of an utterance completes. 1364 * 1365 * @param listener The listener to use. 1366 * 1367 * @return {@link #ERROR} or {@link #SUCCESS}. 1368 * 1369 * @deprecated Use {@link #setOnUtteranceProgressListener(UtteranceProgressListener)} 1370 * instead. 1371 */ 1372 @Deprecated 1373 public int setOnUtteranceCompletedListener(final OnUtteranceCompletedListener listener) { 1374 mUtteranceProgressListener = UtteranceProgressListener.from(listener); 1375 return TextToSpeech.SUCCESS; 1376 } 1377 1378 /** 1379 * Sets the listener that will be notified of various events related to the 1380 * synthesis of a given utterance. 1381 * 1382 * See {@link UtteranceProgressListener} and 1383 * {@link TextToSpeech.Engine#KEY_PARAM_UTTERANCE_ID}. 1384 * 1385 * @param listener the listener to use. 1386 * @return {@link #ERROR} or {@link #SUCCESS} 1387 */ 1388 public int setOnUtteranceProgressListener(UtteranceProgressListener listener) { 1389 mUtteranceProgressListener = listener; 1390 return TextToSpeech.SUCCESS; 1391 } 1392 1393 /** 1394 * Sets the TTS engine to use. 1395 * 1396 * @deprecated This doesn't inform callers when the TTS engine has been 1397 * initialized. {@link #TextToSpeech(Context, OnInitListener, String)} 1398 * can be used with the appropriate engine name. Also, there is no 1399 * guarantee that the engine specified will be loaded. If it isn't 1400 * installed or disabled, the user / system wide defaults will apply. 1401 * 1402 * @param enginePackageName The package name for the synthesis engine (e.g. "com.svox.pico") 1403 * 1404 * @return {@link #ERROR} or {@link #SUCCESS}. 1405 */ 1406 @Deprecated 1407 public int setEngineByPackageName(String enginePackageName) { 1408 mRequestedEngine = enginePackageName; 1409 return initTts(); 1410 } 1411 1412 /** 1413 * Gets the package name of the default speech synthesis engine. 1414 * 1415 * @return Package name of the TTS engine that the user has chosen 1416 * as their default. 1417 */ 1418 public String getDefaultEngine() { 1419 return mEnginesHelper.getDefaultEngine(); 1420 } 1421 1422 /** 1423 * Checks whether the user's settings should override settings requested 1424 * by the calling application. As of the Ice cream sandwich release, 1425 * user settings never forcibly override the app's settings. 1426 */ 1427 public boolean areDefaultsEnforced() { 1428 return false; 1429 } 1430 1431 /** 1432 * Gets a list of all installed TTS engines. 1433 * 1434 * @return A list of engine info objects. The list can be empty, but never {@code null}. 1435 */ 1436 public List<EngineInfo> getEngines() { 1437 return mEnginesHelper.getEngines(); 1438 } 1439 1440 private class Connection implements ServiceConnection { 1441 private ITextToSpeechService mService; 1442 1443 private SetupConnectionAsyncTask mOnSetupConnectionAsyncTask; 1444 1445 private boolean mEstablished; 1446 1447 private final ITextToSpeechCallback.Stub mCallback = new ITextToSpeechCallback.Stub() { 1448 public void onStop(String utteranceId) throws RemoteException { 1449 // do nothing 1450 }; 1451 1452 @Override 1453 public void onFallback(String utteranceId) throws RemoteException { 1454 // do nothing 1455 } 1456 1457 @Override 1458 public void onSuccess(String utteranceId) { 1459 UtteranceProgressListener listener = mUtteranceProgressListener; 1460 if (listener != null) { 1461 listener.onDone(utteranceId); 1462 } 1463 } 1464 1465 @Override 1466 public void onError(String utteranceId, int errorCode) { 1467 UtteranceProgressListener listener = mUtteranceProgressListener; 1468 if (listener != null) { 1469 listener.onError(utteranceId); 1470 } 1471 } 1472 1473 @Override 1474 public void onStart(String utteranceId) { 1475 UtteranceProgressListener listener = mUtteranceProgressListener; 1476 if (listener != null) { 1477 listener.onStart(utteranceId); 1478 } 1479 } 1480 1481 @Override 1482 public void onVoicesInfoChange(List<VoiceInfo> voicesInfo) throws RemoteException { 1483 // Ignore it 1484 } 1485 }; 1486 1487 private class SetupConnectionAsyncTask extends AsyncTask<Void, Void, Integer> { 1488 private final ComponentName mName; 1489 1490 public SetupConnectionAsyncTask(ComponentName name) { 1491 mName = name; 1492 } 1493 1494 @Override 1495 protected Integer doInBackground(Void... params) { 1496 synchronized(mStartLock) { 1497 if (isCancelled()) { 1498 return null; 1499 } 1500 1501 try { 1502 mService.setCallback(getCallerIdentity(), mCallback); 1503 1504 if (mParams.getString(Engine.KEY_PARAM_LANGUAGE) == null) { 1505 String[] defaultLanguage = mService.getClientDefaultLanguage(); 1506 mParams.putString(Engine.KEY_PARAM_LANGUAGE, defaultLanguage[0]); 1507 mParams.putString(Engine.KEY_PARAM_COUNTRY, defaultLanguage[1]); 1508 mParams.putString(Engine.KEY_PARAM_VARIANT, defaultLanguage[2]); 1509 } 1510 1511 Log.i(TAG, "Set up connection to " + mName); 1512 return SUCCESS; 1513 } catch (RemoteException re) { 1514 Log.e(TAG, "Error connecting to service, setCallback() failed"); 1515 return ERROR; 1516 } 1517 } 1518 } 1519 1520 @Override 1521 protected void onPostExecute(Integer result) { 1522 synchronized(mStartLock) { 1523 if (mOnSetupConnectionAsyncTask == this) { 1524 mOnSetupConnectionAsyncTask = null; 1525 } 1526 mEstablished = true; 1527 dispatchOnInit(result); 1528 } 1529 } 1530 } 1531 1532 @Override 1533 public void onServiceConnected(ComponentName name, IBinder service) { 1534 synchronized(mStartLock) { 1535 mConnectingServiceConnection = null; 1536 1537 Log.i(TAG, "Connected to " + name); 1538 1539 if (mOnSetupConnectionAsyncTask != null) { 1540 mOnSetupConnectionAsyncTask.cancel(false); 1541 } 1542 1543 mService = ITextToSpeechService.Stub.asInterface(service); 1544 mServiceConnection = Connection.this; 1545 1546 mEstablished = false; 1547 mOnSetupConnectionAsyncTask = new SetupConnectionAsyncTask(name); 1548 mOnSetupConnectionAsyncTask.execute(); 1549 } 1550 } 1551 1552 public IBinder getCallerIdentity() { 1553 return mCallback; 1554 } 1555 1556 /** 1557 * Clear connection related fields and cancel mOnServiceConnectedAsyncTask if set. 1558 * 1559 * @return true if we cancel mOnSetupConnectionAsyncTask in progress. 1560 */ 1561 private boolean clearServiceConnection() { 1562 synchronized(mStartLock) { 1563 boolean result = false; 1564 if (mOnSetupConnectionAsyncTask != null) { 1565 result = mOnSetupConnectionAsyncTask.cancel(false); 1566 mOnSetupConnectionAsyncTask = null; 1567 } 1568 1569 mService = null; 1570 // If this is the active connection, clear it 1571 if (mServiceConnection == this) { 1572 mServiceConnection = null; 1573 } 1574 return result; 1575 } 1576 } 1577 1578 @Override 1579 public void onServiceDisconnected(ComponentName name) { 1580 Log.i(TAG, "Asked to disconnect from " + name); 1581 if (clearServiceConnection()) { 1582 /* We need to protect against a rare case where engine 1583 * dies just after successful connection - and we process onServiceDisconnected 1584 * before OnServiceConnectedAsyncTask.onPostExecute. onServiceDisconnected cancels 1585 * OnServiceConnectedAsyncTask.onPostExecute and we don't call dispatchOnInit 1586 * with ERROR as argument. 1587 */ 1588 dispatchOnInit(ERROR); 1589 } 1590 } 1591 1592 public void disconnect() { 1593 mContext.unbindService(this); 1594 clearServiceConnection(); 1595 } 1596 1597 public boolean isEstablished() { 1598 return mService != null && mEstablished; 1599 } 1600 1601 public <R> R runAction(Action<R> action, R errorResult, String method, 1602 boolean reconnect, boolean onlyEstablishedConnection) { 1603 synchronized (mStartLock) { 1604 try { 1605 if (mService == null) { 1606 Log.w(TAG, method + " failed: not connected to TTS engine"); 1607 return errorResult; 1608 } 1609 if (onlyEstablishedConnection && !isEstablished()) { 1610 Log.w(TAG, method + " failed: TTS engine connection not fully set up"); 1611 return errorResult; 1612 } 1613 return action.run(mService); 1614 } catch (RemoteException ex) { 1615 Log.e(TAG, method + " failed", ex); 1616 if (reconnect) { 1617 disconnect(); 1618 initTts(); 1619 } 1620 return errorResult; 1621 } 1622 } 1623 } 1624 } 1625 1626 private interface Action<R> { 1627 R run(ITextToSpeechService service) throws RemoteException; 1628 } 1629 1630 /** 1631 * Information about an installed text-to-speech engine. 1632 * 1633 * @see TextToSpeech#getEngines 1634 */ 1635 public static class EngineInfo { 1636 /** 1637 * Engine package name.. 1638 */ 1639 public String name; 1640 /** 1641 * Localized label for the engine. 1642 */ 1643 public String label; 1644 /** 1645 * Icon for the engine. 1646 */ 1647 public int icon; 1648 /** 1649 * Whether this engine is a part of the system 1650 * image. 1651 * 1652 * @hide 1653 */ 1654 public boolean system; 1655 /** 1656 * The priority the engine declares for the the intent filter 1657 * {@code android.intent.action.TTS_SERVICE} 1658 * 1659 * @hide 1660 */ 1661 public int priority; 1662 1663 @Override 1664 public String toString() { 1665 return "EngineInfo{name=" + name + "}"; 1666 } 1667 1668 } 1669 1670 /** 1671 * Limit of length of input string passed to speak and synthesizeToFile. 1672 * 1673 * @see #speak 1674 * @see #synthesizeToFile 1675 */ 1676 public static int getMaxSpeechInputLength() { 1677 return 4000; 1678 } 1679} 1680