TextToSpeech.java revision fc4b2890378eb1b6e0b11d60d703eb6854268064
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 */ 57public class TextToSpeech { 58 59 private static final String TAG = "TextToSpeech"; 60 61 /** 62 * Denotes a successful operation. 63 */ 64 public static final int SUCCESS = 0; 65 /** 66 * Denotes a generic operation failure. 67 */ 68 public static final int ERROR = -1; 69 70 /** 71 * Denotes a stop requested by a client. It's used only on the service side of the API, 72 * client should never expect to see this result code. 73 */ 74 public static final int STOPPED = -2; 75 76 /** 77 * Denotes a failure of a TTS engine to synthesize the given input. 78 */ 79 public static final int ERROR_SYNTHESIS = -3; 80 81 /** 82 * Denotes a failure of a TTS service. 83 */ 84 public static final int ERROR_SERVICE = -4; 85 86 /** 87 * Denotes a failure related to the output (audio device or a file). 88 */ 89 public static final int ERROR_OUTPUT = -5; 90 91 /** 92 * Denotes a failure caused by a network connectivity problems. 93 */ 94 public static final int ERROR_NETWORK = -6; 95 96 /** 97 * Denotes a failure caused by network timeout. 98 */ 99 public static final int ERROR_NETWORK_TIMEOUT = -7; 100 101 /** 102 * Denotes a failure caused by an invalid request. 103 */ 104 public static final int ERROR_INVALID_REQUEST = -8; 105 106 /** 107 * Queue mode where all entries in the playback queue (media to be played 108 * and text to be synthesized) are dropped and replaced by the new entry. 109 * Queues are flushed with respect to a given calling app. Entries in the queue 110 * from other callees are not discarded. 111 */ 112 public static final int QUEUE_FLUSH = 0; 113 /** 114 * Queue mode where the new entry is added at the end of the playback queue. 115 */ 116 public static final int QUEUE_ADD = 1; 117 /** 118 * Queue mode where the entire playback queue is purged. This is different 119 * from {@link #QUEUE_FLUSH} in that all entries are purged, not just entries 120 * from a given caller. 121 * 122 * @hide 123 */ 124 static final int QUEUE_DESTROY = 2; 125 126 /** 127 * Denotes the language is available exactly as specified by the locale. 128 */ 129 public static final int LANG_COUNTRY_VAR_AVAILABLE = 2; 130 131 /** 132 * Denotes the language is available for the language and country specified 133 * by the locale, but not the variant. 134 */ 135 public static final int LANG_COUNTRY_AVAILABLE = 1; 136 137 /** 138 * Denotes the language is available for the language by the locale, 139 * but not the country and variant. 140 */ 141 public static final int LANG_AVAILABLE = 0; 142 143 /** 144 * Denotes the language data is missing. 145 */ 146 public static final int LANG_MISSING_DATA = -1; 147 148 /** 149 * Denotes the language is not supported. 150 */ 151 public static final int LANG_NOT_SUPPORTED = -2; 152 153 /** 154 * Broadcast Action: The TextToSpeech synthesizer has completed processing 155 * of all the text in the speech queue. 156 * 157 * Note that this notifies callers when the <b>engine</b> has finished has 158 * processing text data. Audio playback might not have completed (or even started) 159 * at this point. If you wish to be notified when this happens, see 160 * {@link OnUtteranceCompletedListener}. 161 */ 162 @SdkConstant(SdkConstantType.BROADCAST_INTENT_ACTION) 163 public static final String ACTION_TTS_QUEUE_PROCESSING_COMPLETED = 164 "android.speech.tts.TTS_QUEUE_PROCESSING_COMPLETED"; 165 166 /** 167 * Interface definition of a callback to be invoked indicating the completion of the 168 * TextToSpeech engine initialization. 169 */ 170 public interface OnInitListener { 171 /** 172 * Called to signal the completion of the TextToSpeech engine initialization. 173 * 174 * @param status {@link TextToSpeech#SUCCESS} or {@link TextToSpeech#ERROR}. 175 */ 176 public void onInit(int status); 177 } 178 179 /** 180 * Listener that will be called when the TTS service has 181 * completed synthesizing an utterance. This is only called if the utterance 182 * has an utterance ID (see {@link TextToSpeech.Engine#KEY_PARAM_UTTERANCE_ID}). 183 * 184 * @deprecated Use {@link UtteranceProgressListener} instead. 185 */ 186 @Deprecated 187 public interface OnUtteranceCompletedListener { 188 /** 189 * Called when an utterance has been synthesized. 190 * 191 * @param utteranceId the identifier of the utterance. 192 */ 193 public void onUtteranceCompleted(String utteranceId); 194 } 195 196 /** 197 * Constants and parameter names for controlling text-to-speech. These include: 198 * 199 * <ul> 200 * <li> 201 * Intents to ask engine to install data or check its data and 202 * extras for a TTS engine's check data activity. 203 * </li> 204 * <li> 205 * Keys for the parameters passed with speak commands, e.g. 206 * {@link Engine#KEY_PARAM_UTTERANCE_ID}, {@link Engine#KEY_PARAM_STREAM}. 207 * </li> 208 * <li> 209 * A list of feature strings that engines might support, e.g 210 * {@link Engine#KEY_FEATURE_NETWORK_SYNTHESIS}). These values may be passed in to 211 * {@link TextToSpeech#speak} and {@link TextToSpeech#synthesizeToFile} to modify 212 * engine behaviour. The engine can be queried for the set of features it supports 213 * through {@link TextToSpeech#getFeatures(java.util.Locale)}. 214 * </li> 215 * </ul> 216 */ 217 public class Engine { 218 219 /** 220 * Default speech rate. 221 * @hide 222 */ 223 public static final int DEFAULT_RATE = 100; 224 225 /** 226 * Default pitch. 227 * @hide 228 */ 229 public static final int DEFAULT_PITCH = 100; 230 231 /** 232 * Default volume. 233 * @hide 234 */ 235 public static final float DEFAULT_VOLUME = 1.0f; 236 237 /** 238 * Default pan (centered). 239 * @hide 240 */ 241 public static final float DEFAULT_PAN = 0.0f; 242 243 /** 244 * Default value for {@link Settings.Secure#TTS_USE_DEFAULTS}. 245 * @hide 246 */ 247 public static final int USE_DEFAULTS = 0; // false 248 249 /** 250 * Package name of the default TTS engine. 251 * 252 * @hide 253 * @deprecated No longer in use, the default engine is determined by 254 * the sort order defined in {@link TtsEngines}. Note that 255 * this doesn't "break" anything because there is no guarantee that 256 * the engine specified below is installed on a given build, let 257 * alone be the default. 258 */ 259 @Deprecated 260 public static final String DEFAULT_ENGINE = "com.svox.pico"; 261 262 /** 263 * Default audio stream used when playing synthesized speech. 264 */ 265 public static final int DEFAULT_STREAM = AudioManager.STREAM_MUSIC; 266 267 /** 268 * Indicates success when checking the installation status of the resources used by the 269 * TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 270 */ 271 public static final int CHECK_VOICE_DATA_PASS = 1; 272 273 /** 274 * Indicates failure when checking the installation status of the resources used by the 275 * TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 276 */ 277 public static final int CHECK_VOICE_DATA_FAIL = 0; 278 279 /** 280 * Indicates erroneous data when checking the installation status of the resources used by 281 * the TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 282 * 283 * @deprecated Use CHECK_VOICE_DATA_FAIL instead. 284 */ 285 @Deprecated 286 public static final int CHECK_VOICE_DATA_BAD_DATA = -1; 287 288 /** 289 * Indicates missing resources when checking the installation status of the resources used 290 * by the TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 291 * 292 * @deprecated Use CHECK_VOICE_DATA_FAIL instead. 293 */ 294 @Deprecated 295 public static final int CHECK_VOICE_DATA_MISSING_DATA = -2; 296 297 /** 298 * Indicates missing storage volume when checking the installation status of the resources 299 * used by the TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 300 * 301 * @deprecated Use CHECK_VOICE_DATA_FAIL instead. 302 */ 303 @Deprecated 304 public static final int CHECK_VOICE_DATA_MISSING_VOLUME = -3; 305 306 /** 307 * Intent for starting a TTS service. Services that handle this intent must 308 * extend {@link TextToSpeechService}. Normal applications should not use this intent 309 * directly, instead they should talk to the TTS service using the the methods in this 310 * class. 311 */ 312 @SdkConstant(SdkConstantType.SERVICE_ACTION) 313 public static final String INTENT_ACTION_TTS_SERVICE = 314 "android.intent.action.TTS_SERVICE"; 315 316 /** 317 * Name under which a text to speech engine publishes information about itself. 318 * This meta-data should reference an XML resource containing a 319 * <code><{@link android.R.styleable#TextToSpeechEngine tts-engine}></code> 320 * tag. 321 */ 322 public static final String SERVICE_META_DATA = "android.speech.tts"; 323 324 // intents to ask engine to install data or check its data 325 /** 326 * Activity Action: Triggers the platform TextToSpeech engine to 327 * start the activity that installs the resource files on the device 328 * that are required for TTS to be operational. Since the installation 329 * of the data can be interrupted or declined by the user, the application 330 * shouldn't expect successful installation upon return from that intent, 331 * and if need be, should check installation status with 332 * {@link #ACTION_CHECK_TTS_DATA}. 333 */ 334 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 335 public static final String ACTION_INSTALL_TTS_DATA = 336 "android.speech.tts.engine.INSTALL_TTS_DATA"; 337 338 /** 339 * Broadcast Action: broadcast to signal the change in the list of available 340 * languages or/and their features. 341 */ 342 @SdkConstant(SdkConstantType.BROADCAST_INTENT_ACTION) 343 public static final String ACTION_TTS_DATA_INSTALLED = 344 "android.speech.tts.engine.TTS_DATA_INSTALLED"; 345 346 /** 347 * Activity Action: Starts the activity from the platform TextToSpeech 348 * engine to verify the proper installation and availability of the 349 * resource files on the system. Upon completion, the activity will 350 * return one of the following codes: 351 * {@link #CHECK_VOICE_DATA_PASS}, 352 * {@link #CHECK_VOICE_DATA_FAIL}, 353 * <p> Moreover, the data received in the activity result will contain the following 354 * fields: 355 * <ul> 356 * <li>{@link #EXTRA_AVAILABLE_VOICES} which contains an ArrayList<String> of all the 357 * available voices. The format of each voice is: lang-COUNTRY-variant where COUNTRY and 358 * variant are optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE").</li> 359 * <li>{@link #EXTRA_UNAVAILABLE_VOICES} which contains an ArrayList<String> of all the 360 * unavailable voices (ones that user can install). The format of each voice is: 361 * lang-COUNTRY-variant where COUNTRY and variant are optional (ie, "eng" or 362 * "eng-USA" or "eng-USA-FEMALE").</li> 363 * </ul> 364 */ 365 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 366 public static final String ACTION_CHECK_TTS_DATA = 367 "android.speech.tts.engine.CHECK_TTS_DATA"; 368 369 /** 370 * Activity intent for getting some sample text to use for demonstrating TTS. Specific 371 * locale have to be requested by passing following extra parameters: 372 * <ul> 373 * <li>language</li> 374 * <li>country</li> 375 * <li>variant</li> 376 * </ul> 377 * 378 * Upon completion, the activity result may contain the following fields: 379 * <ul> 380 * <li>{@link #EXTRA_SAMPLE_TEXT} which contains an String with sample text.</li> 381 * </ul> 382 */ 383 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 384 public static final String ACTION_GET_SAMPLE_TEXT = 385 "android.speech.tts.engine.GET_SAMPLE_TEXT"; 386 387 /** 388 * Extra information received with the {@link #ACTION_GET_SAMPLE_TEXT} intent result where 389 * the TextToSpeech engine returns an String with sample text for requested voice 390 */ 391 public static final String EXTRA_SAMPLE_TEXT = "sampleText"; 392 393 394 // extras for a TTS engine's check data activity 395 /** 396 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 397 * the TextToSpeech engine returns an ArrayList<String> of all the available voices. 398 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 399 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 400 */ 401 public static final String EXTRA_AVAILABLE_VOICES = "availableVoices"; 402 403 /** 404 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 405 * the TextToSpeech engine returns an ArrayList<String> of all the unavailable voices. 406 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 407 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 408 */ 409 public static final String EXTRA_UNAVAILABLE_VOICES = "unavailableVoices"; 410 411 /** 412 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 413 * the TextToSpeech engine specifies the path to its resources. 414 * 415 * It may be used by language packages to find out where to put their data. 416 * 417 * @deprecated TTS engine implementation detail, this information has no use for 418 * text-to-speech API client. 419 */ 420 @Deprecated 421 public static final String EXTRA_VOICE_DATA_ROOT_DIRECTORY = "dataRoot"; 422 423 /** 424 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 425 * the TextToSpeech engine specifies the file names of its resources under the 426 * resource path. 427 * 428 * @deprecated TTS engine implementation detail, this information has no use for 429 * text-to-speech API client. 430 */ 431 @Deprecated 432 public static final String EXTRA_VOICE_DATA_FILES = "dataFiles"; 433 434 /** 435 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 436 * the TextToSpeech engine specifies the locale associated with each resource file. 437 * 438 * @deprecated TTS engine implementation detail, this information has no use for 439 * text-to-speech API client. 440 */ 441 @Deprecated 442 public static final String EXTRA_VOICE_DATA_FILES_INFO = "dataFilesInfo"; 443 444 /** 445 * Extra information sent with the {@link #ACTION_CHECK_TTS_DATA} intent where the 446 * caller indicates to the TextToSpeech engine which specific sets of voice data to 447 * check for by sending an ArrayList<String> of the voices that are of interest. 448 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 449 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 450 * 451 * @deprecated Redundant functionality, checking for existence of specific sets of voice 452 * data can be done on client side. 453 */ 454 @Deprecated 455 public static final String EXTRA_CHECK_VOICE_DATA_FOR = "checkVoiceDataFor"; 456 457 // extras for a TTS engine's data installation 458 /** 459 * Extra information received with the {@link #ACTION_TTS_DATA_INSTALLED} intent result. 460 * It indicates whether the data files for the synthesis engine were successfully 461 * installed. The installation was initiated with the {@link #ACTION_INSTALL_TTS_DATA} 462 * intent. The possible values for this extra are 463 * {@link TextToSpeech#SUCCESS} and {@link TextToSpeech#ERROR}. 464 * 465 * @deprecated No longer in use. If client ise interested in information about what 466 * changed, is should send ACTION_CHECK_TTS_DATA intent to discover available voices. 467 */ 468 @Deprecated 469 public static final String EXTRA_TTS_DATA_INSTALLED = "dataInstalled"; 470 471 // keys for the parameters passed with speak commands. Hidden keys are used internally 472 // to maintain engine state for each TextToSpeech instance. 473 /** 474 * @hide 475 */ 476 public static final String KEY_PARAM_RATE = "rate"; 477 478 /** 479 * @hide 480 */ 481 public static final String KEY_PARAM_LANGUAGE = "language"; 482 483 /** 484 * @hide 485 */ 486 public static final String KEY_PARAM_COUNTRY = "country"; 487 488 /** 489 * @hide 490 */ 491 public static final String KEY_PARAM_VARIANT = "variant"; 492 493 /** 494 * @hide 495 */ 496 public static final String KEY_PARAM_ENGINE = "engine"; 497 498 /** 499 * @hide 500 */ 501 public static final String KEY_PARAM_PITCH = "pitch"; 502 503 /** 504 * Parameter key to specify the audio stream type to be used when speaking text 505 * or playing back a file. The value should be one of the STREAM_ constants 506 * defined in {@link AudioManager}. 507 * 508 * @see TextToSpeech#speak(String, int, HashMap) 509 * @see TextToSpeech#playEarcon(String, int, HashMap) 510 */ 511 public static final String KEY_PARAM_STREAM = "streamType"; 512 513 /** 514 * Parameter key to identify an utterance in the 515 * {@link TextToSpeech.OnUtteranceCompletedListener} after text has been 516 * spoken, a file has been played back or a silence duration has elapsed. 517 * 518 * @see TextToSpeech#speak(String, int, HashMap) 519 * @see TextToSpeech#playEarcon(String, int, HashMap) 520 * @see TextToSpeech#synthesizeToFile(String, HashMap, String) 521 */ 522 public static final String KEY_PARAM_UTTERANCE_ID = "utteranceId"; 523 524 /** 525 * Parameter key to specify the speech volume relative to the current stream type 526 * volume used when speaking text. Volume is specified as a float ranging from 0 to 1 527 * where 0 is silence, and 1 is the maximum volume (the default behavior). 528 * 529 * @see TextToSpeech#speak(String, int, HashMap) 530 * @see TextToSpeech#playEarcon(String, int, HashMap) 531 */ 532 public static final String KEY_PARAM_VOLUME = "volume"; 533 534 /** 535 * Parameter key to specify how the speech is panned from left to right when speaking text. 536 * Pan is specified as a float ranging from -1 to +1 where -1 maps to a hard-left pan, 537 * 0 to center (the default behavior), and +1 to hard-right. 538 * 539 * @see TextToSpeech#speak(String, int, HashMap) 540 * @see TextToSpeech#playEarcon(String, int, HashMap) 541 */ 542 public static final String KEY_PARAM_PAN = "pan"; 543 544 /** 545 * Feature key for network synthesis. See {@link TextToSpeech#getFeatures(Locale)} 546 * for a description of how feature keys work. If set (and supported by the engine 547 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must 548 * use network based synthesis. 549 * 550 * @see TextToSpeech#speak(String, int, java.util.HashMap) 551 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 552 * @see TextToSpeech#getFeatures(java.util.Locale) 553 */ 554 public static final String KEY_FEATURE_NETWORK_SYNTHESIS = "networkTts"; 555 556 /** 557 * Feature key for embedded synthesis. See {@link TextToSpeech#getFeatures(Locale)} 558 * for a description of how feature keys work. If set and supported by the engine 559 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must synthesize 560 * text on-device (without making network requests). 561 * 562 * @see TextToSpeech#speak(String, int, java.util.HashMap) 563 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 564 * @see TextToSpeech#getFeatures(java.util.Locale) 565 */ 566 public static final String KEY_FEATURE_EMBEDDED_SYNTHESIS = "embeddedTts"; 567 } 568 569 private final Context mContext; 570 private Connection mConnectingServiceConnection; 571 private Connection mServiceConnection; 572 private OnInitListener mInitListener; 573 // Written from an unspecified application thread, read from 574 // a binder thread. 575 private volatile UtteranceProgressListener mUtteranceProgressListener; 576 private final Object mStartLock = new Object(); 577 578 private String mRequestedEngine; 579 // Whether to initialize this TTS object with the default engine, 580 // if the requested engine is not available. Valid only if mRequestedEngine 581 // is not null. Used only for testing, though potentially useful API wise 582 // too. 583 private final boolean mUseFallback; 584 private final Map<String, Uri> mEarcons; 585 private final Map<String, Uri> mUtterances; 586 private final Bundle mParams = new Bundle(); 587 private final TtsEngines mEnginesHelper; 588 private final String mPackageName; 589 private volatile String mCurrentEngine = null; 590 591 /** 592 * The constructor for the TextToSpeech class, using the default TTS engine. 593 * This will also initialize the associated TextToSpeech engine if it isn't already running. 594 * 595 * @param context 596 * The context this instance is running in. 597 * @param listener 598 * The {@link TextToSpeech.OnInitListener} that will be called when the 599 * TextToSpeech engine has initialized. In a case of a failure the listener 600 * may be called immediately, before TextToSpeech instance is fully constructed. 601 */ 602 public TextToSpeech(Context context, OnInitListener listener) { 603 this(context, listener, null); 604 } 605 606 /** 607 * The constructor for the TextToSpeech class, using the given TTS engine. 608 * This will also initialize the associated TextToSpeech engine if it isn't already running. 609 * 610 * @param context 611 * The context this instance is running in. 612 * @param listener 613 * The {@link TextToSpeech.OnInitListener} that will be called when the 614 * TextToSpeech engine has initialized. In a case of a failure the listener 615 * may be called immediately, before TextToSpeech instance is fully constructed. 616 * @param engine Package name of the TTS engine to use. 617 */ 618 public TextToSpeech(Context context, OnInitListener listener, String engine) { 619 this(context, listener, engine, null, true); 620 } 621 622 /** 623 * Used by the framework to instantiate TextToSpeech objects with a supplied 624 * package name, instead of using {@link android.content.Context#getPackageName()} 625 * 626 * @hide 627 */ 628 public TextToSpeech(Context context, OnInitListener listener, String engine, 629 String packageName, boolean useFallback) { 630 mContext = context; 631 mInitListener = listener; 632 mRequestedEngine = engine; 633 mUseFallback = useFallback; 634 635 mEarcons = new HashMap<String, Uri>(); 636 mUtterances = new HashMap<String, Uri>(); 637 mUtteranceProgressListener = null; 638 639 mEnginesHelper = new TtsEngines(mContext); 640 if (packageName != null) { 641 mPackageName = packageName; 642 } else { 643 mPackageName = mContext.getPackageName(); 644 } 645 initTts(); 646 } 647 648 private <R> R runActionNoReconnect(Action<R> action, R errorResult, String method, 649 boolean onlyEstablishedConnection) { 650 return runAction(action, errorResult, method, false, onlyEstablishedConnection); 651 } 652 653 private <R> R runAction(Action<R> action, R errorResult, String method) { 654 return runAction(action, errorResult, method, true, true); 655 } 656 657 private <R> R runAction(Action<R> action, R errorResult, String method, 658 boolean reconnect, boolean onlyEstablishedConnection) { 659 synchronized (mStartLock) { 660 if (mServiceConnection == null) { 661 Log.w(TAG, method + " failed: not bound to TTS engine"); 662 return errorResult; 663 } 664 return mServiceConnection.runAction(action, errorResult, method, reconnect, 665 onlyEstablishedConnection); 666 } 667 } 668 669 private int initTts() { 670 // Step 1: Try connecting to the engine that was requested. 671 if (mRequestedEngine != null) { 672 if (mEnginesHelper.isEngineInstalled(mRequestedEngine)) { 673 if (connectToEngine(mRequestedEngine)) { 674 mCurrentEngine = mRequestedEngine; 675 return SUCCESS; 676 } else if (!mUseFallback) { 677 mCurrentEngine = null; 678 dispatchOnInit(ERROR); 679 return ERROR; 680 } 681 } else if (!mUseFallback) { 682 Log.i(TAG, "Requested engine not installed: " + mRequestedEngine); 683 mCurrentEngine = null; 684 dispatchOnInit(ERROR); 685 return ERROR; 686 } 687 } 688 689 // Step 2: Try connecting to the user's default engine. 690 final String defaultEngine = getDefaultEngine(); 691 if (defaultEngine != null && !defaultEngine.equals(mRequestedEngine)) { 692 if (connectToEngine(defaultEngine)) { 693 mCurrentEngine = defaultEngine; 694 return SUCCESS; 695 } 696 } 697 698 // Step 3: Try connecting to the highest ranked engine in the 699 // system. 700 final String highestRanked = mEnginesHelper.getHighestRankedEngineName(); 701 if (highestRanked != null && !highestRanked.equals(mRequestedEngine) && 702 !highestRanked.equals(defaultEngine)) { 703 if (connectToEngine(highestRanked)) { 704 mCurrentEngine = highestRanked; 705 return SUCCESS; 706 } 707 } 708 709 // NOTE: The API currently does not allow the caller to query whether 710 // they are actually connected to any engine. This might fail for various 711 // reasons like if the user disables all her TTS engines. 712 713 mCurrentEngine = null; 714 dispatchOnInit(ERROR); 715 return ERROR; 716 } 717 718 private boolean connectToEngine(String engine) { 719 Connection connection = new Connection(); 720 Intent intent = new Intent(Engine.INTENT_ACTION_TTS_SERVICE); 721 intent.setPackage(engine); 722 boolean bound = mContext.bindService(intent, connection, Context.BIND_AUTO_CREATE); 723 if (!bound) { 724 Log.e(TAG, "Failed to bind to " + engine); 725 return false; 726 } else { 727 Log.i(TAG, "Sucessfully bound to " + engine); 728 mConnectingServiceConnection = connection; 729 return true; 730 } 731 } 732 733 private void dispatchOnInit(int result) { 734 synchronized (mStartLock) { 735 if (mInitListener != null) { 736 mInitListener.onInit(result); 737 mInitListener = null; 738 } 739 } 740 } 741 742 private IBinder getCallerIdentity() { 743 return mServiceConnection.getCallerIdentity(); 744 } 745 746 /** 747 * Releases the resources used by the TextToSpeech engine. 748 * It is good practice for instance to call this method in the onDestroy() method of an Activity 749 * so the TextToSpeech engine can be cleanly stopped. 750 */ 751 public void shutdown() { 752 // Special case, we are asked to shutdown connection that did finalize its connection. 753 synchronized (mStartLock) { 754 if (mConnectingServiceConnection != null) { 755 mContext.unbindService(mConnectingServiceConnection); 756 mConnectingServiceConnection = null; 757 return; 758 } 759 } 760 761 // Post connection case 762 runActionNoReconnect(new Action<Void>() { 763 @Override 764 public Void run(ITextToSpeechService service) throws RemoteException { 765 service.setCallback(getCallerIdentity(), null); 766 service.stop(getCallerIdentity()); 767 mServiceConnection.disconnect(); 768 // Context#unbindService does not result in a call to 769 // ServiceConnection#onServiceDisconnected. As a result, the 770 // service ends up being destroyed (if there are no other open 771 // connections to it) but the process lives on and the 772 // ServiceConnection continues to refer to the destroyed service. 773 // 774 // This leads to tons of log spam about SynthThread being dead. 775 mServiceConnection = null; 776 mCurrentEngine = null; 777 return null; 778 } 779 }, null, "shutdown", false); 780 } 781 782 /** 783 * Adds a mapping between a string of text and a sound resource in a 784 * package. After a call to this method, subsequent calls to 785 * {@link #speak(String, int, HashMap)} will play the specified sound resource 786 * if it is available, or synthesize the text it is missing. 787 * 788 * @param text 789 * The string of text. Example: <code>"south_south_east"</code> 790 * 791 * @param packagename 792 * Pass the packagename of the application that contains the 793 * resource. If the resource is in your own application (this is 794 * the most common case), then put the packagename of your 795 * application here.<br/> 796 * Example: <b>"com.google.marvin.compass"</b><br/> 797 * The packagename can be found in the AndroidManifest.xml of 798 * your application. 799 * <p> 800 * <code><manifest xmlns:android="..." 801 * package="<b>com.google.marvin.compass</b>"></code> 802 * </p> 803 * 804 * @param resourceId 805 * Example: <code>R.raw.south_south_east</code> 806 * 807 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 808 */ 809 public int addSpeech(String text, String packagename, int resourceId) { 810 synchronized (mStartLock) { 811 mUtterances.put(text, makeResourceUri(packagename, resourceId)); 812 return SUCCESS; 813 } 814 } 815 816 /** 817 * Adds a mapping between a string of text and a sound file. Using this, it 818 * is possible to add custom pronounciations for a string of text. 819 * After a call to this method, subsequent calls to {@link #speak(String, int, HashMap)} 820 * will play the specified sound resource if it is available, or synthesize the text it is 821 * missing. 822 * 823 * @param text 824 * The string of text. Example: <code>"south_south_east"</code> 825 * @param filename 826 * The full path to the sound file (for example: 827 * "/sdcard/mysounds/hello.wav") 828 * 829 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 830 */ 831 public int addSpeech(String text, String filename) { 832 synchronized (mStartLock) { 833 mUtterances.put(text, Uri.parse(filename)); 834 return SUCCESS; 835 } 836 } 837 838 839 /** 840 * Adds a mapping between a string of text and a sound resource in a 841 * package. Use this to add custom earcons. 842 * 843 * @see #playEarcon(String, int, HashMap) 844 * 845 * @param earcon The name of the earcon. 846 * Example: <code>"[tick]"</code><br/> 847 * 848 * @param packagename 849 * the package name of the application that contains the 850 * resource. This can for instance be the package name of your own application. 851 * Example: <b>"com.google.marvin.compass"</b><br/> 852 * The package name can be found in the AndroidManifest.xml of 853 * the application containing the resource. 854 * <p> 855 * <code><manifest xmlns:android="..." 856 * package="<b>com.google.marvin.compass</b>"></code> 857 * </p> 858 * 859 * @param resourceId 860 * Example: <code>R.raw.tick_snd</code> 861 * 862 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 863 */ 864 public int addEarcon(String earcon, String packagename, int resourceId) { 865 synchronized(mStartLock) { 866 mEarcons.put(earcon, makeResourceUri(packagename, resourceId)); 867 return SUCCESS; 868 } 869 } 870 871 /** 872 * Adds a mapping between a string of text and a sound file. 873 * Use this to add custom earcons. 874 * 875 * @see #playEarcon(String, int, HashMap) 876 * 877 * @param earcon 878 * The name of the earcon. 879 * Example: <code>"[tick]"</code> 880 * @param filename 881 * The full path to the sound file (for example: 882 * "/sdcard/mysounds/tick.wav") 883 * 884 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 885 */ 886 public int addEarcon(String earcon, String filename) { 887 synchronized(mStartLock) { 888 mEarcons.put(earcon, Uri.parse(filename)); 889 return SUCCESS; 890 } 891 } 892 893 private Uri makeResourceUri(String packageName, int resourceId) { 894 return new Uri.Builder() 895 .scheme(ContentResolver.SCHEME_ANDROID_RESOURCE) 896 .encodedAuthority(packageName) 897 .appendEncodedPath(String.valueOf(resourceId)) 898 .build(); 899 } 900 901 /** 902 * Speaks the string using the specified queuing strategy and speech parameters. 903 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 904 * requests and then returns. The synthesis might not have finished (or even started!) at the 905 * time when this method returns. In order to reliably detect errors during synthesis, 906 * we recommend setting an utterance progress listener (see 907 * {@link #setOnUtteranceProgressListener}) and using the 908 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 909 * 910 * @param text The string of text to be spoken. No longer than 911 * {@link #getMaxSpeechInputLength()} characters. 912 * @param queueMode The queuing strategy to use, {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 913 * @param params Parameters for the request. Can be null. 914 * Supported parameter names: 915 * {@link Engine#KEY_PARAM_STREAM}, 916 * {@link Engine#KEY_PARAM_UTTERANCE_ID}, 917 * {@link Engine#KEY_PARAM_VOLUME}, 918 * {@link Engine#KEY_PARAM_PAN}. 919 * Engine specific parameters may be passed in but the parameter keys 920 * must be prefixed by the name of the engine they are intended for. For example 921 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 922 * engine named "com.svox.pico" if it is being used. 923 * 924 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the speak operation. 925 */ 926 public int speak(final String text, final int queueMode, final HashMap<String, String> params) { 927 return runAction(new Action<Integer>() { 928 @Override 929 public Integer run(ITextToSpeechService service) throws RemoteException { 930 Uri utteranceUri = mUtterances.get(text); 931 if (utteranceUri != null) { 932 return service.playAudio(getCallerIdentity(), utteranceUri, queueMode, 933 getParams(params)); 934 } else { 935 return service.speak(getCallerIdentity(), text, queueMode, getParams(params)); 936 } 937 } 938 }, ERROR, "speak"); 939 } 940 941 /** 942 * Plays the earcon using the specified queueing mode and parameters. 943 * The earcon must already have been added with {@link #addEarcon(String, String)} or 944 * {@link #addEarcon(String, String, int)}. 945 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 946 * requests and then returns. The synthesis might not have finished (or even started!) at the 947 * time when this method returns. In order to reliably detect errors during synthesis, 948 * we recommend setting an utterance progress listener (see 949 * {@link #setOnUtteranceProgressListener}) and using the 950 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 951 * 952 * @param earcon The earcon that should be played 953 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 954 * @param params Parameters for the request. Can be null. 955 * Supported parameter names: 956 * {@link Engine#KEY_PARAM_STREAM}, 957 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 958 * Engine specific parameters may be passed in but the parameter keys 959 * must be prefixed by the name of the engine they are intended for. For example 960 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 961 * engine named "com.svox.pico" if it is being used. 962 * 963 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playEarcon operation. 964 */ 965 public int playEarcon(final String earcon, final int queueMode, 966 final HashMap<String, String> params) { 967 return runAction(new Action<Integer>() { 968 @Override 969 public Integer run(ITextToSpeechService service) throws RemoteException { 970 Uri earconUri = mEarcons.get(earcon); 971 if (earconUri == null) { 972 return ERROR; 973 } 974 return service.playAudio(getCallerIdentity(), earconUri, queueMode, 975 getParams(params)); 976 } 977 }, ERROR, "playEarcon"); 978 } 979 980 /** 981 * Plays silence for the specified amount of time using the specified 982 * queue mode. 983 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 984 * requests and then returns. The synthesis might not have finished (or even started!) at the 985 * time when this method returns. In order to reliably detect errors during synthesis, 986 * we recommend setting an utterance progress listener (see 987 * {@link #setOnUtteranceProgressListener}) and using the 988 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 989 * 990 * @param durationInMs The duration of the silence. 991 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 992 * @param params Parameters for the request. Can be null. 993 * Supported parameter names: 994 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 995 * Engine specific parameters may be passed in but the parameter keys 996 * must be prefixed by the name of the engine they are intended for. For example 997 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 998 * engine named "com.svox.pico" if it is being used. 999 * 1000 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playSilence operation. 1001 */ 1002 public int playSilence(final long durationInMs, final int queueMode, 1003 final HashMap<String, String> params) { 1004 return runAction(new Action<Integer>() { 1005 @Override 1006 public Integer run(ITextToSpeechService service) throws RemoteException { 1007 return service.playSilence(getCallerIdentity(), durationInMs, queueMode, 1008 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1009 } 1010 }, ERROR, "playSilence"); 1011 } 1012 1013 /** 1014 * Queries the engine for the set of features it supports for a given locale. 1015 * Features can either be framework defined, e.g. 1016 * {@link TextToSpeech.Engine#KEY_FEATURE_NETWORK_SYNTHESIS} or engine specific. 1017 * Engine specific keys must be prefixed by the name of the engine they 1018 * are intended for. These keys can be used as parameters to 1019 * {@link TextToSpeech#speak(String, int, java.util.HashMap)} and 1020 * {@link TextToSpeech#synthesizeToFile(String, java.util.HashMap, String)}. 1021 * 1022 * Features are boolean flags, and their values in the synthesis parameters 1023 * must be behave as per {@link Boolean#parseBoolean(String)}. 1024 * 1025 * @param locale The locale to query features for. 1026 * @return Set instance. May return {@code null} on error. 1027 */ 1028 public Set<String> getFeatures(final Locale locale) { 1029 return runAction(new Action<Set<String>>() { 1030 @Override 1031 public Set<String> run(ITextToSpeechService service) throws RemoteException { 1032 String[] features = null; 1033 try { 1034 features = service.getFeaturesForLanguage( 1035 locale.getISO3Language(), locale.getISO3Country(), locale.getVariant()); 1036 } catch(MissingResourceException e) { 1037 Log.w(TAG, "Couldn't retrieve 3 letter ISO 639-2/T language and/or ISO 3166 " + 1038 "country code for locale: " + locale, e); 1039 return null; 1040 } 1041 1042 if (features != null) { 1043 final Set<String> featureSet = new HashSet<String>(); 1044 Collections.addAll(featureSet, features); 1045 return featureSet; 1046 } 1047 return null; 1048 } 1049 }, null, "getFeatures"); 1050 } 1051 1052 /** 1053 * Checks whether the TTS engine is busy speaking. Note that a speech item is 1054 * considered complete once it's audio data has been sent to the audio mixer, or 1055 * written to a file. There might be a finite lag between this point, and when 1056 * the audio hardware completes playback. 1057 * 1058 * @return {@code true} if the TTS engine is speaking. 1059 */ 1060 public boolean isSpeaking() { 1061 return runAction(new Action<Boolean>() { 1062 @Override 1063 public Boolean run(ITextToSpeechService service) throws RemoteException { 1064 return service.isSpeaking(); 1065 } 1066 }, false, "isSpeaking"); 1067 } 1068 1069 /** 1070 * Interrupts the current utterance (whether played or rendered to file) and discards other 1071 * utterances in the queue. 1072 * 1073 * @return {@link #ERROR} or {@link #SUCCESS}. 1074 */ 1075 public int stop() { 1076 return runAction(new Action<Integer>() { 1077 @Override 1078 public Integer run(ITextToSpeechService service) throws RemoteException { 1079 return service.stop(getCallerIdentity()); 1080 } 1081 }, ERROR, "stop"); 1082 } 1083 1084 /** 1085 * Sets the speech rate. 1086 * 1087 * This has no effect on any pre-recorded speech. 1088 * 1089 * @param speechRate Speech rate. {@code 1.0} is the normal speech rate, 1090 * lower values slow down the speech ({@code 0.5} is half the normal speech rate), 1091 * greater values accelerate it ({@code 2.0} is twice the normal speech rate). 1092 * 1093 * @return {@link #ERROR} or {@link #SUCCESS}. 1094 */ 1095 public int setSpeechRate(float speechRate) { 1096 if (speechRate > 0.0f) { 1097 int intRate = (int)(speechRate * 100); 1098 if (intRate > 0) { 1099 synchronized (mStartLock) { 1100 mParams.putInt(Engine.KEY_PARAM_RATE, intRate); 1101 } 1102 return SUCCESS; 1103 } 1104 } 1105 return ERROR; 1106 } 1107 1108 /** 1109 * Sets the speech pitch for the TextToSpeech engine. 1110 * 1111 * This has no effect on any pre-recorded speech. 1112 * 1113 * @param pitch Speech pitch. {@code 1.0} is the normal pitch, 1114 * lower values lower the tone of the synthesized voice, 1115 * greater values increase it. 1116 * 1117 * @return {@link #ERROR} or {@link #SUCCESS}. 1118 */ 1119 public int setPitch(float pitch) { 1120 if (pitch > 0.0f) { 1121 int intPitch = (int)(pitch * 100); 1122 if (intPitch > 0) { 1123 synchronized (mStartLock) { 1124 mParams.putInt(Engine.KEY_PARAM_PITCH, intPitch); 1125 } 1126 return SUCCESS; 1127 } 1128 } 1129 return ERROR; 1130 } 1131 1132 /** 1133 * @return the engine currently in use by this TextToSpeech instance. 1134 * @hide 1135 */ 1136 public String getCurrentEngine() { 1137 return mCurrentEngine; 1138 } 1139 1140 /** 1141 * Returns a Locale instance describing the language currently being used as the default 1142 * Text-to-speech language. 1143 * 1144 * @return language, country (if any) and variant (if any) used by the client stored in a 1145 * Locale instance, or {@code null} on error. 1146 */ 1147 public Locale getDefaultLanguage() { 1148 return runAction(new Action<Locale>() { 1149 @Override 1150 public Locale run(ITextToSpeechService service) throws RemoteException { 1151 String[] defaultLanguage = service.getClientDefaultLanguage(); 1152 1153 return new Locale(defaultLanguage[0], defaultLanguage[1], defaultLanguage[2]); 1154 } 1155 }, null, "getDefaultLanguage"); 1156 } 1157 1158 /** 1159 * Sets the text-to-speech language. 1160 * The TTS engine will try to use the closest match to the specified 1161 * language as represented by the Locale, but there is no guarantee that the exact same Locale 1162 * will be used. Use {@link #isLanguageAvailable(Locale)} to check the level of support 1163 * before choosing the language to use for the next utterances. 1164 * 1165 * @param loc The locale describing the language to be used. 1166 * 1167 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1168 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1169 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1170 */ 1171 public int setLanguage(final Locale loc) { 1172 return runAction(new Action<Integer>() { 1173 @Override 1174 public Integer run(ITextToSpeechService service) throws RemoteException { 1175 if (loc == null) { 1176 return LANG_NOT_SUPPORTED; 1177 } 1178 String language = null, country = null; 1179 try { 1180 language = loc.getISO3Language(); 1181 } catch (MissingResourceException e) { 1182 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1183 return LANG_NOT_SUPPORTED; 1184 } 1185 1186 try { 1187 country = loc.getISO3Country(); 1188 } catch (MissingResourceException e) { 1189 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1190 return LANG_NOT_SUPPORTED; 1191 } 1192 1193 String variant = loc.getVariant(); 1194 1195 // Check if the language, country, variant are available, and cache 1196 // the available parts. 1197 // Note that the language is not actually set here, instead it is cached so it 1198 // will be associated with all upcoming utterances. 1199 1200 int result = service.loadLanguage(getCallerIdentity(), language, country, variant); 1201 if (result >= LANG_AVAILABLE){ 1202 if (result < LANG_COUNTRY_VAR_AVAILABLE) { 1203 variant = ""; 1204 if (result < LANG_COUNTRY_AVAILABLE) { 1205 country = ""; 1206 } 1207 } 1208 mParams.putString(Engine.KEY_PARAM_LANGUAGE, language); 1209 mParams.putString(Engine.KEY_PARAM_COUNTRY, country); 1210 mParams.putString(Engine.KEY_PARAM_VARIANT, variant); 1211 } 1212 return result; 1213 } 1214 }, LANG_NOT_SUPPORTED, "setLanguage"); 1215 } 1216 1217 /** 1218 * Returns a Locale instance describing the language currently being used for synthesis 1219 * requests sent to the TextToSpeech engine. 1220 * 1221 * In Android 4.2 and before (API <= 17) this function returns the language that is currently 1222 * being used by the TTS engine. That is the last language set by this or any other 1223 * client by a {@link TextToSpeech#setLanguage} call to the same engine. 1224 * 1225 * In Android versions after 4.2 this function returns the language that is currently being 1226 * used for the synthesis requests sent from this client. That is the last language set 1227 * by a {@link TextToSpeech#setLanguage} call on this instance. 1228 * 1229 * @return language, country (if any) and variant (if any) used by the client stored in a 1230 * Locale instance, or {@code null} on error. 1231 */ 1232 public Locale getLanguage() { 1233 return runAction(new Action<Locale>() { 1234 @Override 1235 public Locale run(ITextToSpeechService service) { 1236 /* No service call, but we're accessing mParams, hence need for 1237 wrapping it as an Action instance */ 1238 String lang = mParams.getString(Engine.KEY_PARAM_LANGUAGE, ""); 1239 String country = mParams.getString(Engine.KEY_PARAM_COUNTRY, ""); 1240 String variant = mParams.getString(Engine.KEY_PARAM_VARIANT, ""); 1241 return new Locale(lang, country, variant); 1242 } 1243 }, null, "getLanguage"); 1244 } 1245 1246 /** 1247 * Checks if the specified language as represented by the Locale is available and supported. 1248 * 1249 * @param loc The Locale describing the language to be used. 1250 * 1251 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1252 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1253 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1254 */ 1255 public int isLanguageAvailable(final Locale loc) { 1256 return runAction(new Action<Integer>() { 1257 @Override 1258 public Integer run(ITextToSpeechService service) throws RemoteException { 1259 String language = null, country = null; 1260 1261 try { 1262 language = loc.getISO3Language(); 1263 } catch (MissingResourceException e) { 1264 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1265 return LANG_NOT_SUPPORTED; 1266 } 1267 1268 try { 1269 country = loc.getISO3Country(); 1270 } catch (MissingResourceException e) { 1271 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1272 return LANG_NOT_SUPPORTED; 1273 } 1274 1275 return service.isLanguageAvailable(language, country, loc.getVariant()); 1276 } 1277 }, LANG_NOT_SUPPORTED, "isLanguageAvailable"); 1278 } 1279 1280 /** 1281 * Synthesizes the given text to a file using the specified parameters. 1282 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1283 * requests and then returns. The synthesis might not have finished (or even started!) at the 1284 * time when this method returns. In order to reliably detect errors during synthesis, 1285 * we recommend setting an utterance progress listener (see 1286 * {@link #setOnUtteranceProgressListener}) and using the 1287 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1288 * 1289 * @param text The text that should be synthesized. No longer than 1290 * {@link #getMaxSpeechInputLength()} characters. 1291 * @param params Parameters for the request. Can be null. 1292 * Supported parameter names: 1293 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1294 * Engine specific parameters may be passed in but the parameter keys 1295 * must be prefixed by the name of the engine they are intended for. For example 1296 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1297 * engine named "com.svox.pico" if it is being used. 1298 * @param filename Absolute file filename to write the generated audio data to.It should be 1299 * something like "/sdcard/myappsounds/mysound.wav". 1300 * 1301 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the synthesizeToFile operation. 1302 */ 1303 public int synthesizeToFile(final String text, final HashMap<String, String> params, 1304 final String filename) { 1305 return runAction(new Action<Integer>() { 1306 @Override 1307 public Integer run(ITextToSpeechService service) throws RemoteException { 1308 ParcelFileDescriptor fileDescriptor; 1309 int returnValue; 1310 try { 1311 File file = new File(filename); 1312 if(file.exists() && !file.canWrite()) { 1313 Log.e(TAG, "Can't write to " + filename); 1314 return ERROR; 1315 } 1316 fileDescriptor = ParcelFileDescriptor.open(file, 1317 ParcelFileDescriptor.MODE_WRITE_ONLY | 1318 ParcelFileDescriptor.MODE_CREATE | 1319 ParcelFileDescriptor.MODE_TRUNCATE); 1320 returnValue = service.synthesizeToFileDescriptor(getCallerIdentity(), text, 1321 fileDescriptor, getParams(params)); 1322 fileDescriptor.close(); 1323 return returnValue; 1324 } catch (FileNotFoundException e) { 1325 Log.e(TAG, "Opening file " + filename + " failed", e); 1326 return ERROR; 1327 } catch (IOException e) { 1328 Log.e(TAG, "Closing file " + filename + " failed", e); 1329 return ERROR; 1330 } 1331 } 1332 }, ERROR, "synthesizeToFile"); 1333 } 1334 1335 private Bundle getParams(HashMap<String, String> params) { 1336 if (params != null && !params.isEmpty()) { 1337 Bundle bundle = new Bundle(mParams); 1338 copyIntParam(bundle, params, Engine.KEY_PARAM_STREAM); 1339 copyStringParam(bundle, params, Engine.KEY_PARAM_UTTERANCE_ID); 1340 copyFloatParam(bundle, params, Engine.KEY_PARAM_VOLUME); 1341 copyFloatParam(bundle, params, Engine.KEY_PARAM_PAN); 1342 1343 // Copy feature strings defined by the framework. 1344 copyStringParam(bundle, params, Engine.KEY_FEATURE_NETWORK_SYNTHESIS); 1345 copyStringParam(bundle, params, Engine.KEY_FEATURE_EMBEDDED_SYNTHESIS); 1346 1347 // Copy over all parameters that start with the name of the 1348 // engine that we are currently connected to. The engine is 1349 // free to interpret them as it chooses. 1350 if (!TextUtils.isEmpty(mCurrentEngine)) { 1351 for (Map.Entry<String, String> entry : params.entrySet()) { 1352 final String key = entry.getKey(); 1353 if (key != null && key.startsWith(mCurrentEngine)) { 1354 bundle.putString(key, entry.getValue()); 1355 } 1356 } 1357 } 1358 1359 return bundle; 1360 } else { 1361 return mParams; 1362 } 1363 } 1364 1365 private void copyStringParam(Bundle bundle, HashMap<String, String> params, String key) { 1366 String value = params.get(key); 1367 if (value != null) { 1368 bundle.putString(key, value); 1369 } 1370 } 1371 1372 private void copyIntParam(Bundle bundle, HashMap<String, String> params, String key) { 1373 String valueString = params.get(key); 1374 if (!TextUtils.isEmpty(valueString)) { 1375 try { 1376 int value = Integer.parseInt(valueString); 1377 bundle.putInt(key, value); 1378 } catch (NumberFormatException ex) { 1379 // don't set the value in the bundle 1380 } 1381 } 1382 } 1383 1384 private void copyFloatParam(Bundle bundle, HashMap<String, String> params, String key) { 1385 String valueString = params.get(key); 1386 if (!TextUtils.isEmpty(valueString)) { 1387 try { 1388 float value = Float.parseFloat(valueString); 1389 bundle.putFloat(key, value); 1390 } catch (NumberFormatException ex) { 1391 // don't set the value in the bundle 1392 } 1393 } 1394 } 1395 1396 /** 1397 * Sets the listener that will be notified when synthesis of an utterance completes. 1398 * 1399 * @param listener The listener to use. 1400 * 1401 * @return {@link #ERROR} or {@link #SUCCESS}. 1402 * 1403 * @deprecated Use {@link #setOnUtteranceProgressListener(UtteranceProgressListener)} 1404 * instead. 1405 */ 1406 @Deprecated 1407 public int setOnUtteranceCompletedListener(final OnUtteranceCompletedListener listener) { 1408 mUtteranceProgressListener = UtteranceProgressListener.from(listener); 1409 return TextToSpeech.SUCCESS; 1410 } 1411 1412 /** 1413 * Sets the listener that will be notified of various events related to the 1414 * synthesis of a given utterance. 1415 * 1416 * See {@link UtteranceProgressListener} and 1417 * {@link TextToSpeech.Engine#KEY_PARAM_UTTERANCE_ID}. 1418 * 1419 * @param listener the listener to use. 1420 * @return {@link #ERROR} or {@link #SUCCESS} 1421 */ 1422 public int setOnUtteranceProgressListener(UtteranceProgressListener listener) { 1423 mUtteranceProgressListener = listener; 1424 return TextToSpeech.SUCCESS; 1425 } 1426 1427 /** 1428 * Sets the TTS engine to use. 1429 * 1430 * @deprecated This doesn't inform callers when the TTS engine has been 1431 * initialized. {@link #TextToSpeech(Context, OnInitListener, String)} 1432 * can be used with the appropriate engine name. Also, there is no 1433 * guarantee that the engine specified will be loaded. If it isn't 1434 * installed or disabled, the user / system wide defaults will apply. 1435 * 1436 * @param enginePackageName The package name for the synthesis engine (e.g. "com.svox.pico") 1437 * 1438 * @return {@link #ERROR} or {@link #SUCCESS}. 1439 */ 1440 @Deprecated 1441 public int setEngineByPackageName(String enginePackageName) { 1442 mRequestedEngine = enginePackageName; 1443 return initTts(); 1444 } 1445 1446 /** 1447 * Gets the package name of the default speech synthesis engine. 1448 * 1449 * @return Package name of the TTS engine that the user has chosen 1450 * as their default. 1451 */ 1452 public String getDefaultEngine() { 1453 return mEnginesHelper.getDefaultEngine(); 1454 } 1455 1456 /** 1457 * Checks whether the user's settings should override settings requested 1458 * by the calling application. As of the Ice cream sandwich release, 1459 * user settings never forcibly override the app's settings. 1460 */ 1461 public boolean areDefaultsEnforced() { 1462 return false; 1463 } 1464 1465 /** 1466 * Gets a list of all installed TTS engines. 1467 * 1468 * @return A list of engine info objects. The list can be empty, but never {@code null}. 1469 */ 1470 public List<EngineInfo> getEngines() { 1471 return mEnginesHelper.getEngines(); 1472 } 1473 1474 private class Connection implements ServiceConnection { 1475 private ITextToSpeechService mService; 1476 1477 private SetupConnectionAsyncTask mOnSetupConnectionAsyncTask; 1478 1479 private boolean mEstablished; 1480 1481 private final ITextToSpeechCallback.Stub mCallback = new ITextToSpeechCallback.Stub() { 1482 public void onStop(String utteranceId) throws RemoteException { 1483 // do nothing 1484 }; 1485 1486 @Override 1487 public void onFallback(String utteranceId) throws RemoteException { 1488 // do nothing 1489 } 1490 1491 @Override 1492 public void onSuccess(String utteranceId) { 1493 UtteranceProgressListener listener = mUtteranceProgressListener; 1494 if (listener != null) { 1495 listener.onDone(utteranceId); 1496 } 1497 } 1498 1499 @Override 1500 public void onError(String utteranceId, int errorCode) { 1501 UtteranceProgressListener listener = mUtteranceProgressListener; 1502 if (listener != null) { 1503 listener.onError(utteranceId); 1504 } 1505 } 1506 1507 @Override 1508 public void onStart(String utteranceId) { 1509 UtteranceProgressListener listener = mUtteranceProgressListener; 1510 if (listener != null) { 1511 listener.onStart(utteranceId); 1512 } 1513 } 1514 }; 1515 1516 private class SetupConnectionAsyncTask extends AsyncTask<Void, Void, Integer> { 1517 private final ComponentName mName; 1518 1519 public SetupConnectionAsyncTask(ComponentName name) { 1520 mName = name; 1521 } 1522 1523 @Override 1524 protected Integer doInBackground(Void... params) { 1525 synchronized(mStartLock) { 1526 if (isCancelled()) { 1527 return null; 1528 } 1529 1530 try { 1531 mService.setCallback(getCallerIdentity(), mCallback); 1532 1533 if (mParams.getString(Engine.KEY_PARAM_LANGUAGE) == null) { 1534 String[] defaultLanguage = mService.getClientDefaultLanguage(); 1535 mParams.putString(Engine.KEY_PARAM_LANGUAGE, defaultLanguage[0]); 1536 mParams.putString(Engine.KEY_PARAM_COUNTRY, defaultLanguage[1]); 1537 mParams.putString(Engine.KEY_PARAM_VARIANT, defaultLanguage[2]); 1538 } 1539 1540 Log.i(TAG, "Set up connection to " + mName); 1541 return SUCCESS; 1542 } catch (RemoteException re) { 1543 Log.e(TAG, "Error connecting to service, setCallback() failed"); 1544 return ERROR; 1545 } 1546 } 1547 } 1548 1549 @Override 1550 protected void onPostExecute(Integer result) { 1551 synchronized(mStartLock) { 1552 if (mOnSetupConnectionAsyncTask == this) { 1553 mOnSetupConnectionAsyncTask = null; 1554 } 1555 mEstablished = true; 1556 dispatchOnInit(result); 1557 } 1558 } 1559 } 1560 1561 @Override 1562 public void onServiceConnected(ComponentName name, IBinder service) { 1563 synchronized(mStartLock) { 1564 mConnectingServiceConnection = null; 1565 1566 Log.i(TAG, "Connected to " + name); 1567 1568 if (mOnSetupConnectionAsyncTask != null) { 1569 mOnSetupConnectionAsyncTask.cancel(false); 1570 } 1571 1572 mService = ITextToSpeechService.Stub.asInterface(service); 1573 mServiceConnection = Connection.this; 1574 1575 mEstablished = false; 1576 mOnSetupConnectionAsyncTask = new SetupConnectionAsyncTask(name); 1577 mOnSetupConnectionAsyncTask.execute(); 1578 } 1579 } 1580 1581 public IBinder getCallerIdentity() { 1582 return mCallback; 1583 } 1584 1585 /** 1586 * Clear connection related fields and cancel mOnServiceConnectedAsyncTask if set. 1587 * 1588 * @return true if we cancel mOnSetupConnectionAsyncTask in progress. 1589 */ 1590 private boolean clearServiceConnection() { 1591 synchronized(mStartLock) { 1592 boolean result = false; 1593 if (mOnSetupConnectionAsyncTask != null) { 1594 result = mOnSetupConnectionAsyncTask.cancel(false); 1595 mOnSetupConnectionAsyncTask = null; 1596 } 1597 1598 mService = null; 1599 // If this is the active connection, clear it 1600 if (mServiceConnection == this) { 1601 mServiceConnection = null; 1602 } 1603 return result; 1604 } 1605 } 1606 1607 @Override 1608 public void onServiceDisconnected(ComponentName name) { 1609 Log.i(TAG, "Asked to disconnect from " + name); 1610 if (clearServiceConnection()) { 1611 /* We need to protect against a rare case where engine 1612 * dies just after successful connection - and we process onServiceDisconnected 1613 * before OnServiceConnectedAsyncTask.onPostExecute. onServiceDisconnected cancels 1614 * OnServiceConnectedAsyncTask.onPostExecute and we don't call dispatchOnInit 1615 * with ERROR as argument. 1616 */ 1617 dispatchOnInit(ERROR); 1618 } 1619 } 1620 1621 public void disconnect() { 1622 mContext.unbindService(this); 1623 clearServiceConnection(); 1624 } 1625 1626 public boolean isEstablished() { 1627 return mService != null && mEstablished; 1628 } 1629 1630 public <R> R runAction(Action<R> action, R errorResult, String method, 1631 boolean reconnect, boolean onlyEstablishedConnection) { 1632 synchronized (mStartLock) { 1633 try { 1634 if (mService == null) { 1635 Log.w(TAG, method + " failed: not connected to TTS engine"); 1636 return errorResult; 1637 } 1638 if (onlyEstablishedConnection && !isEstablished()) { 1639 Log.w(TAG, method + " failed: TTS engine connection not fully set up"); 1640 return errorResult; 1641 } 1642 return action.run(mService); 1643 } catch (RemoteException ex) { 1644 Log.e(TAG, method + " failed", ex); 1645 if (reconnect) { 1646 disconnect(); 1647 initTts(); 1648 } 1649 return errorResult; 1650 } 1651 } 1652 } 1653 } 1654 1655 private interface Action<R> { 1656 R run(ITextToSpeechService service) throws RemoteException; 1657 } 1658 1659 /** 1660 * Information about an installed text-to-speech engine. 1661 * 1662 * @see TextToSpeech#getEngines 1663 */ 1664 public static class EngineInfo { 1665 /** 1666 * Engine package name.. 1667 */ 1668 public String name; 1669 /** 1670 * Localized label for the engine. 1671 */ 1672 public String label; 1673 /** 1674 * Icon for the engine. 1675 */ 1676 public int icon; 1677 /** 1678 * Whether this engine is a part of the system 1679 * image. 1680 * 1681 * @hide 1682 */ 1683 public boolean system; 1684 /** 1685 * The priority the engine declares for the the intent filter 1686 * {@code android.intent.action.TTS_SERVICE} 1687 * 1688 * @hide 1689 */ 1690 public int priority; 1691 1692 @Override 1693 public String toString() { 1694 return "EngineInfo{name=" + name + "}"; 1695 } 1696 1697 } 1698 1699 /** 1700 * Limit of length of input string passed to speak and synthesizeToFile. 1701 * 1702 * @see #speak 1703 * @see #synthesizeToFile 1704 */ 1705 public static int getMaxSpeechInputLength() { 1706 return 4000; 1707 } 1708} 1709