TextToSpeech.java revision 7b9c912f536925ac6ec43935d6e97506851b33d6
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.RawRes; 19import android.annotation.SdkConstant; 20import android.annotation.SdkConstant.SdkConstantType; 21import android.content.ComponentName; 22import android.content.ContentResolver; 23import android.content.Context; 24import android.content.Intent; 25import android.content.ServiceConnection; 26import android.media.AudioAttributes; 27import android.media.AudioManager; 28import android.net.Uri; 29import android.os.AsyncTask; 30import android.os.Bundle; 31import android.os.IBinder; 32import android.os.ParcelFileDescriptor; 33import android.os.RemoteException; 34import android.provider.Settings; 35import android.text.TextUtils; 36import android.util.Log; 37 38import java.io.File; 39import java.io.FileNotFoundException; 40import java.io.IOException; 41import java.util.ArrayList; 42import java.util.Collections; 43import java.util.HashMap; 44import java.util.HashSet; 45import java.util.List; 46import java.util.Locale; 47import java.util.Map; 48import java.util.MissingResourceException; 49import java.util.Set; 50 51/** 52 * 53 * Synthesizes speech from text for immediate playback or to create a sound file. 54 * <p>A TextToSpeech instance can only be used to synthesize text once it has completed its 55 * initialization. Implement the {@link TextToSpeech.OnInitListener} to be 56 * notified of the completion of the initialization.<br> 57 * When you are done using the TextToSpeech instance, call the {@link #shutdown()} method 58 * to release the native resources used by the TextToSpeech engine. 59 */ 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 * Denotes a stop requested by a client. It's used only on the service side of the API, 75 * client should never expect to see this result code. 76 */ 77 public static final int STOPPED = -2; 78 79 /** 80 * Denotes a failure of a TTS engine to synthesize the given input. 81 */ 82 public static final int ERROR_SYNTHESIS = -3; 83 84 /** 85 * Denotes a failure of a TTS service. 86 */ 87 public static final int ERROR_SERVICE = -4; 88 89 /** 90 * Denotes a failure related to the output (audio device or a file). 91 */ 92 public static final int ERROR_OUTPUT = -5; 93 94 /** 95 * Denotes a failure caused by a network connectivity problems. 96 */ 97 public static final int ERROR_NETWORK = -6; 98 99 /** 100 * Denotes a failure caused by network timeout. 101 */ 102 public static final int ERROR_NETWORK_TIMEOUT = -7; 103 104 /** 105 * Denotes a failure caused by an invalid request. 106 */ 107 public static final int ERROR_INVALID_REQUEST = -8; 108 109 /** 110 * Denotes a failure caused by an unfinished download of the voice data. 111 * @see Engine#KEY_FEATURE_NOT_INSTALLED 112 */ 113 public static final int ERROR_NOT_INSTALLED_YET = -9; 114 115 /** 116 * Queue mode where all entries in the playback queue (media to be played 117 * and text to be synthesized) are dropped and replaced by the new entry. 118 * Queues are flushed with respect to a given calling app. Entries in the queue 119 * from other callees are not discarded. 120 */ 121 public static final int QUEUE_FLUSH = 0; 122 /** 123 * Queue mode where the new entry is added at the end of the playback queue. 124 */ 125 public static final int QUEUE_ADD = 1; 126 /** 127 * Queue mode where the entire playback queue is purged. This is different 128 * from {@link #QUEUE_FLUSH} in that all entries are purged, not just entries 129 * from a given caller. 130 * 131 * @hide 132 */ 133 static final int QUEUE_DESTROY = 2; 134 135 /** 136 * Denotes the language is available exactly as specified by the locale. 137 */ 138 public static final int LANG_COUNTRY_VAR_AVAILABLE = 2; 139 140 /** 141 * Denotes the language is available for the language and country specified 142 * by the locale, but not the variant. 143 */ 144 public static final int LANG_COUNTRY_AVAILABLE = 1; 145 146 /** 147 * Denotes the language is available for the language by the locale, 148 * but not the country and variant. 149 */ 150 public static final int LANG_AVAILABLE = 0; 151 152 /** 153 * Denotes the language data is missing. 154 */ 155 public static final int LANG_MISSING_DATA = -1; 156 157 /** 158 * Denotes the language is not supported. 159 */ 160 public static final int LANG_NOT_SUPPORTED = -2; 161 162 /** 163 * Broadcast Action: The TextToSpeech synthesizer has completed processing 164 * of all the text in the speech queue. 165 * 166 * Note that this notifies callers when the <b>engine</b> has finished has 167 * processing text data. Audio playback might not have completed (or even started) 168 * at this point. If you wish to be notified when this happens, see 169 * {@link OnUtteranceCompletedListener}. 170 */ 171 @SdkConstant(SdkConstantType.BROADCAST_INTENT_ACTION) 172 public static final String ACTION_TTS_QUEUE_PROCESSING_COMPLETED = 173 "android.speech.tts.TTS_QUEUE_PROCESSING_COMPLETED"; 174 175 /** 176 * Interface definition of a callback to be invoked indicating the completion of the 177 * TextToSpeech engine initialization. 178 */ 179 public interface OnInitListener { 180 /** 181 * Called to signal the completion of the TextToSpeech engine initialization. 182 * 183 * @param status {@link TextToSpeech#SUCCESS} or {@link TextToSpeech#ERROR}. 184 */ 185 public void onInit(int status); 186 } 187 188 /** 189 * Listener that will be called when the TTS service has 190 * completed synthesizing an utterance. This is only called if the utterance 191 * has an utterance ID (see {@link TextToSpeech.Engine#KEY_PARAM_UTTERANCE_ID}). 192 * 193 * @deprecated Use {@link UtteranceProgressListener} instead. 194 */ 195 @Deprecated 196 public interface OnUtteranceCompletedListener { 197 /** 198 * Called when an utterance has been synthesized. 199 * 200 * @param utteranceId the identifier of the utterance. 201 */ 202 public void onUtteranceCompleted(String utteranceId); 203 } 204 205 /** 206 * Constants and parameter names for controlling text-to-speech. These include: 207 * 208 * <ul> 209 * <li> 210 * Intents to ask engine to install data or check its data and 211 * extras for a TTS engine's check data activity. 212 * </li> 213 * <li> 214 * Keys for the parameters passed with speak commands, e.g. 215 * {@link Engine#KEY_PARAM_UTTERANCE_ID}, {@link Engine#KEY_PARAM_STREAM}. 216 * </li> 217 * <li> 218 * A list of feature strings that engines might support, e.g 219 * {@link Engine#KEY_FEATURE_NETWORK_SYNTHESIS}). These values may be passed in to 220 * {@link TextToSpeech#speak} and {@link TextToSpeech#synthesizeToFile} to modify 221 * engine behaviour. The engine can be queried for the set of features it supports 222 * through {@link TextToSpeech#getFeatures(java.util.Locale)}. 223 * </li> 224 * </ul> 225 */ 226 public class Engine { 227 228 /** 229 * Default speech rate. 230 * @hide 231 */ 232 public static final int DEFAULT_RATE = 100; 233 234 /** 235 * Default pitch. 236 * @hide 237 */ 238 public static final int DEFAULT_PITCH = 100; 239 240 /** 241 * Default volume. 242 * @hide 243 */ 244 public static final float DEFAULT_VOLUME = 1.0f; 245 246 /** 247 * Default pan (centered). 248 * @hide 249 */ 250 public static final float DEFAULT_PAN = 0.0f; 251 252 /** 253 * Default value for {@link Settings.Secure#TTS_USE_DEFAULTS}. 254 * @hide 255 */ 256 public static final int USE_DEFAULTS = 0; // false 257 258 /** 259 * Package name of the default TTS engine. 260 * 261 * @hide 262 * @deprecated No longer in use, the default engine is determined by 263 * the sort order defined in {@link TtsEngines}. Note that 264 * this doesn't "break" anything because there is no guarantee that 265 * the engine specified below is installed on a given build, let 266 * alone be the default. 267 */ 268 @Deprecated 269 public static final String DEFAULT_ENGINE = "com.svox.pico"; 270 271 /** 272 * Default audio stream used when playing synthesized speech. 273 */ 274 public static final int DEFAULT_STREAM = AudioManager.STREAM_MUSIC; 275 276 /** 277 * Indicates success when checking the installation status of the resources used by the 278 * TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 279 */ 280 public static final int CHECK_VOICE_DATA_PASS = 1; 281 282 /** 283 * Indicates failure when checking the installation status of the resources used by the 284 * TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 285 */ 286 public static final int CHECK_VOICE_DATA_FAIL = 0; 287 288 /** 289 * Indicates erroneous data when checking the installation status of the resources used by 290 * 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_BAD_DATA = -1; 296 297 /** 298 * Indicates missing resources when checking the installation status of the resources used 299 * 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_DATA = -2; 305 306 /** 307 * Indicates missing storage volume when checking the installation status of the resources 308 * used by the TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 309 * 310 * @deprecated Use CHECK_VOICE_DATA_FAIL instead. 311 */ 312 @Deprecated 313 public static final int CHECK_VOICE_DATA_MISSING_VOLUME = -3; 314 315 /** 316 * Intent for starting a TTS service. Services that handle this intent must 317 * extend {@link TextToSpeechService}. Normal applications should not use this intent 318 * directly, instead they should talk to the TTS service using the the methods in this 319 * class. 320 */ 321 @SdkConstant(SdkConstantType.SERVICE_ACTION) 322 public static final String INTENT_ACTION_TTS_SERVICE = 323 "android.intent.action.TTS_SERVICE"; 324 325 /** 326 * Name under which a text to speech engine publishes information about itself. 327 * This meta-data should reference an XML resource containing a 328 * <code><{@link android.R.styleable#TextToSpeechEngine tts-engine}></code> 329 * tag. 330 */ 331 public static final String SERVICE_META_DATA = "android.speech.tts"; 332 333 // intents to ask engine to install data or check its data 334 /** 335 * Activity Action: Triggers the platform TextToSpeech engine to 336 * start the activity that installs the resource files on the device 337 * that are required for TTS to be operational. Since the installation 338 * of the data can be interrupted or declined by the user, the application 339 * shouldn't expect successful installation upon return from that intent, 340 * and if need be, should check installation status with 341 * {@link #ACTION_CHECK_TTS_DATA}. 342 */ 343 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 344 public static final String ACTION_INSTALL_TTS_DATA = 345 "android.speech.tts.engine.INSTALL_TTS_DATA"; 346 347 /** 348 * Broadcast Action: broadcast to signal the change in the list of available 349 * languages or/and their features. 350 */ 351 @SdkConstant(SdkConstantType.BROADCAST_INTENT_ACTION) 352 public static final String ACTION_TTS_DATA_INSTALLED = 353 "android.speech.tts.engine.TTS_DATA_INSTALLED"; 354 355 /** 356 * Activity Action: Starts the activity from the platform TextToSpeech 357 * engine to verify the proper installation and availability of the 358 * resource files on the system. Upon completion, the activity will 359 * return one of the following codes: 360 * {@link #CHECK_VOICE_DATA_PASS}, 361 * {@link #CHECK_VOICE_DATA_FAIL}, 362 * <p> Moreover, the data received in the activity result will contain the following 363 * fields: 364 * <ul> 365 * <li>{@link #EXTRA_AVAILABLE_VOICES} which contains an ArrayList<String> of all the 366 * available voices. The format of each voice is: lang-COUNTRY-variant where COUNTRY and 367 * variant are optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE").</li> 368 * <li>{@link #EXTRA_UNAVAILABLE_VOICES} which contains an ArrayList<String> of all the 369 * unavailable voices (ones that user can install). The format of each voice is: 370 * lang-COUNTRY-variant where COUNTRY and variant are optional (ie, "eng" or 371 * "eng-USA" or "eng-USA-FEMALE").</li> 372 * </ul> 373 */ 374 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 375 public static final String ACTION_CHECK_TTS_DATA = 376 "android.speech.tts.engine.CHECK_TTS_DATA"; 377 378 /** 379 * Activity intent for getting some sample text to use for demonstrating TTS. Specific 380 * locale have to be requested by passing following extra parameters: 381 * <ul> 382 * <li>language</li> 383 * <li>country</li> 384 * <li>variant</li> 385 * </ul> 386 * 387 * Upon completion, the activity result may contain the following fields: 388 * <ul> 389 * <li>{@link #EXTRA_SAMPLE_TEXT} which contains an String with sample text.</li> 390 * </ul> 391 */ 392 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 393 public static final String ACTION_GET_SAMPLE_TEXT = 394 "android.speech.tts.engine.GET_SAMPLE_TEXT"; 395 396 /** 397 * Extra information received with the {@link #ACTION_GET_SAMPLE_TEXT} intent result where 398 * the TextToSpeech engine returns an String with sample text for requested voice 399 */ 400 public static final String EXTRA_SAMPLE_TEXT = "sampleText"; 401 402 403 // extras for a TTS engine's check data activity 404 /** 405 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 406 * the TextToSpeech engine returns an ArrayList<String> of all the available voices. 407 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 408 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 409 */ 410 public static final String EXTRA_AVAILABLE_VOICES = "availableVoices"; 411 412 /** 413 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 414 * the TextToSpeech engine returns an ArrayList<String> of all the unavailable voices. 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 public static final String EXTRA_UNAVAILABLE_VOICES = "unavailableVoices"; 419 420 /** 421 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 422 * the TextToSpeech engine specifies the path to its resources. 423 * 424 * It may be used by language packages to find out where to put their data. 425 * 426 * @deprecated TTS engine implementation detail, this information has no use for 427 * text-to-speech API client. 428 */ 429 @Deprecated 430 public static final String EXTRA_VOICE_DATA_ROOT_DIRECTORY = "dataRoot"; 431 432 /** 433 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 434 * the TextToSpeech engine specifies the file names of its resources under the 435 * resource path. 436 * 437 * @deprecated TTS engine implementation detail, this information has no use for 438 * text-to-speech API client. 439 */ 440 @Deprecated 441 public static final String EXTRA_VOICE_DATA_FILES = "dataFiles"; 442 443 /** 444 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 445 * the TextToSpeech engine specifies the locale associated with each resource file. 446 * 447 * @deprecated TTS engine implementation detail, this information has no use for 448 * text-to-speech API client. 449 */ 450 @Deprecated 451 public static final String EXTRA_VOICE_DATA_FILES_INFO = "dataFilesInfo"; 452 453 /** 454 * Extra information sent with the {@link #ACTION_CHECK_TTS_DATA} intent where the 455 * caller indicates to the TextToSpeech engine which specific sets of voice data to 456 * check for by sending an ArrayList<String> of the voices that are of interest. 457 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 458 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 459 * 460 * @deprecated Redundant functionality, checking for existence of specific sets of voice 461 * data can be done on client side. 462 */ 463 @Deprecated 464 public static final String EXTRA_CHECK_VOICE_DATA_FOR = "checkVoiceDataFor"; 465 466 // extras for a TTS engine's data installation 467 /** 468 * Extra information received with the {@link #ACTION_TTS_DATA_INSTALLED} intent result. 469 * It indicates whether the data files for the synthesis engine were successfully 470 * installed. The installation was initiated with the {@link #ACTION_INSTALL_TTS_DATA} 471 * intent. The possible values for this extra are 472 * {@link TextToSpeech#SUCCESS} and {@link TextToSpeech#ERROR}. 473 * 474 * @deprecated No longer in use. If client ise interested in information about what 475 * changed, is should send ACTION_CHECK_TTS_DATA intent to discover available voices. 476 */ 477 @Deprecated 478 public static final String EXTRA_TTS_DATA_INSTALLED = "dataInstalled"; 479 480 // keys for the parameters passed with speak commands. Hidden keys are used internally 481 // to maintain engine state for each TextToSpeech instance. 482 /** 483 * @hide 484 */ 485 public static final String KEY_PARAM_RATE = "rate"; 486 487 /** 488 * @hide 489 */ 490 public static final String KEY_PARAM_VOICE_NAME = "voiceName"; 491 492 /** 493 * @hide 494 */ 495 public static final String KEY_PARAM_LANGUAGE = "language"; 496 497 /** 498 * @hide 499 */ 500 public static final String KEY_PARAM_COUNTRY = "country"; 501 502 /** 503 * @hide 504 */ 505 public static final String KEY_PARAM_VARIANT = "variant"; 506 507 /** 508 * @hide 509 */ 510 public static final String KEY_PARAM_ENGINE = "engine"; 511 512 /** 513 * @hide 514 */ 515 public static final String KEY_PARAM_PITCH = "pitch"; 516 517 /** 518 * Parameter key to specify the audio stream type to be used when speaking text 519 * or playing back a file. The value should be one of the STREAM_ constants 520 * defined in {@link AudioManager}. 521 * 522 * @see TextToSpeech#speak(String, int, HashMap) 523 * @see TextToSpeech#playEarcon(String, int, HashMap) 524 */ 525 public static final String KEY_PARAM_STREAM = "streamType"; 526 527 /** 528 * Parameter key to specify the audio attributes to be used when 529 * speaking text or playing back a file. The value should be set 530 * using {@link TextToSpeech#setAudioAttributes(AudioAttributes)}. 531 * 532 * @see TextToSpeech#speak(String, int, HashMap) 533 * @see TextToSpeech#playEarcon(String, int, HashMap) 534 * @hide 535 */ 536 public static final String KEY_PARAM_AUDIO_ATTRIBUTES = "audioAttributes"; 537 538 /** 539 * Parameter key to identify an utterance in the 540 * {@link TextToSpeech.OnUtteranceCompletedListener} after text has been 541 * spoken, a file has been played back or a silence duration has elapsed. 542 * 543 * @see TextToSpeech#speak(String, int, HashMap) 544 * @see TextToSpeech#playEarcon(String, int, HashMap) 545 * @see TextToSpeech#synthesizeToFile(String, HashMap, String) 546 */ 547 public static final String KEY_PARAM_UTTERANCE_ID = "utteranceId"; 548 549 /** 550 * Parameter key to specify the speech volume relative to the current stream type 551 * volume used when speaking text. Volume is specified as a float ranging from 0 to 1 552 * where 0 is silence, and 1 is the maximum volume (the default behavior). 553 * 554 * @see TextToSpeech#speak(String, int, HashMap) 555 * @see TextToSpeech#playEarcon(String, int, HashMap) 556 */ 557 public static final String KEY_PARAM_VOLUME = "volume"; 558 559 /** 560 * Parameter key to specify how the speech is panned from left to right when speaking text. 561 * Pan is specified as a float ranging from -1 to +1 where -1 maps to a hard-left pan, 562 * 0 to center (the default behavior), and +1 to hard-right. 563 * 564 * @see TextToSpeech#speak(String, int, HashMap) 565 * @see TextToSpeech#playEarcon(String, int, HashMap) 566 */ 567 public static final String KEY_PARAM_PAN = "pan"; 568 569 /** 570 * Feature key for network synthesis. See {@link TextToSpeech#getFeatures(Locale)} 571 * for a description of how feature keys work. If set (and supported by the engine 572 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must 573 * use network based synthesis. 574 * 575 * @see TextToSpeech#speak(String, int, java.util.HashMap) 576 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 577 * @see TextToSpeech#getFeatures(java.util.Locale) 578 * 579 * @deprecated Starting from API level 21, to select network synthesis, call 580 * ({@link TextToSpeech#getVoices()}, find a suitable network voice 581 * ({@link Voice#isNetworkConnectionRequired()}) and pass it 582 * to {@link TextToSpeech#setVoice(Voice)}). 583 */ 584 @Deprecated 585 public static final String KEY_FEATURE_NETWORK_SYNTHESIS = "networkTts"; 586 587 /** 588 * Feature key for embedded synthesis. See {@link TextToSpeech#getFeatures(Locale)} 589 * for a description of how feature keys work. If set and supported by the engine 590 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must synthesize 591 * text on-device (without making network requests). 592 * 593 * @see TextToSpeech#speak(String, int, java.util.HashMap) 594 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 595 * @see TextToSpeech#getFeatures(java.util.Locale) 596 597 * @deprecated Starting from API level 21, to select embedded synthesis, call 598 * ({@link TextToSpeech#getVoices()}, find a suitable embedded voice 599 * ({@link Voice#isNetworkConnectionRequired()}) and pass it 600 * to {@link TextToSpeech#setVoice(Voice)}). 601 */ 602 @Deprecated 603 public static final String KEY_FEATURE_EMBEDDED_SYNTHESIS = "embeddedTts"; 604 605 /** 606 * Parameter key to specify an audio session identifier (obtained from 607 * {@link AudioManager#generateAudioSessionId()}) that will be used by the request audio 608 * output. It can be used to associate one of the {@link android.media.audiofx.AudioEffect} 609 * objects with the synthesis (or earcon) output. 610 * 611 * @see TextToSpeech#speak(String, int, HashMap) 612 * @see TextToSpeech#playEarcon(String, int, HashMap) 613 */ 614 public static final String KEY_PARAM_SESSION_ID = "sessionId"; 615 616 /** 617 * Feature key that indicates that the voice may need to download additional data to be fully 618 * functional. The download will be triggered by calling 619 * {@link TextToSpeech#setVoice(Voice)} or {@link TextToSpeech#setLanguage(Locale)}. 620 * Until download is complete, each synthesis request will either report 621 * {@link TextToSpeech#ERROR_NOT_INSTALLED_YET} error, or use a different voice to synthesize 622 * the request. This feature should NOT be used as a key of a request parameter. 623 * 624 * @see TextToSpeech#getFeatures(java.util.Locale) 625 * @see Voice#getFeatures() 626 */ 627 public static final String KEY_FEATURE_NOT_INSTALLED = "notInstalled"; 628 629 /** 630 * Feature key that indicate that a network timeout can be set for the request. If set and 631 * supported as per {@link TextToSpeech#getFeatures(Locale)} or {@link Voice#getFeatures()}, 632 * it can be used as request parameter to set the maximum allowed time for a single 633 * request attempt, in milliseconds, before synthesis fails. When used as a key of 634 * a request parameter, its value should be a string with an integer value. 635 * 636 * @see TextToSpeech#getFeatures(java.util.Locale) 637 * @see Voice#getFeatures() 638 */ 639 public static final String KEY_FEATURE_NETWORK_TIMEOUT_MS = "networkTimeoutMs"; 640 641 /** 642 * Feature key that indicates that network request retries count can be set for the request. 643 * If set and supported as per {@link TextToSpeech#getFeatures(Locale)} or 644 * {@link Voice#getFeatures()}, it can be used as a request parameter to set the 645 * number of network request retries that are attempted in case of failure. When used as 646 * a key of a request parameter, its value should be a string with an integer value. 647 * 648 * @see TextToSpeech#getFeatures(java.util.Locale) 649 * @see Voice#getFeatures() 650 */ 651 public static final String KEY_FEATURE_NETWORK_RETRIES_COUNT = "networkRetriesCount"; 652 } 653 654 private final Context mContext; 655 private Connection mConnectingServiceConnection; 656 private Connection mServiceConnection; 657 private OnInitListener mInitListener; 658 // Written from an unspecified application thread, read from 659 // a binder thread. 660 private volatile UtteranceProgressListener mUtteranceProgressListener; 661 private final Object mStartLock = new Object(); 662 663 private String mRequestedEngine; 664 // Whether to initialize this TTS object with the default engine, 665 // if the requested engine is not available. Valid only if mRequestedEngine 666 // is not null. Used only for testing, though potentially useful API wise 667 // too. 668 private final boolean mUseFallback; 669 private final Map<String, Uri> mEarcons; 670 private final Map<CharSequence, Uri> mUtterances; 671 private final Bundle mParams = new Bundle(); 672 private final TtsEngines mEnginesHelper; 673 private volatile String mCurrentEngine = null; 674 675 /** 676 * The constructor for the TextToSpeech class, using the default TTS engine. 677 * This will also initialize the associated TextToSpeech engine if it isn't already running. 678 * 679 * @param context 680 * The context this instance is running in. 681 * @param listener 682 * The {@link TextToSpeech.OnInitListener} that will be called when the 683 * TextToSpeech engine has initialized. In a case of a failure the listener 684 * may be called immediately, before TextToSpeech instance is fully constructed. 685 */ 686 public TextToSpeech(Context context, OnInitListener listener) { 687 this(context, listener, null); 688 } 689 690 /** 691 * The constructor for the TextToSpeech class, using the given TTS engine. 692 * This will also initialize the associated TextToSpeech engine if it isn't already running. 693 * 694 * @param context 695 * The context this instance is running in. 696 * @param listener 697 * The {@link TextToSpeech.OnInitListener} that will be called when the 698 * TextToSpeech engine has initialized. In a case of a failure the listener 699 * may be called immediately, before TextToSpeech instance is fully constructed. 700 * @param engine Package name of the TTS engine to use. 701 */ 702 public TextToSpeech(Context context, OnInitListener listener, String engine) { 703 this(context, listener, engine, null, true); 704 } 705 706 /** 707 * Used by the framework to instantiate TextToSpeech objects with a supplied 708 * package name, instead of using {@link android.content.Context#getPackageName()} 709 * 710 * @hide 711 */ 712 public TextToSpeech(Context context, OnInitListener listener, String engine, 713 String packageName, boolean useFallback) { 714 mContext = context; 715 mInitListener = listener; 716 mRequestedEngine = engine; 717 mUseFallback = useFallback; 718 719 mEarcons = new HashMap<String, Uri>(); 720 mUtterances = new HashMap<CharSequence, Uri>(); 721 mUtteranceProgressListener = null; 722 723 mEnginesHelper = new TtsEngines(mContext); 724 initTts(); 725 } 726 727 private <R> R runActionNoReconnect(Action<R> action, R errorResult, String method, 728 boolean onlyEstablishedConnection) { 729 return runAction(action, errorResult, method, false, onlyEstablishedConnection); 730 } 731 732 private <R> R runAction(Action<R> action, R errorResult, String method) { 733 return runAction(action, errorResult, method, true, true); 734 } 735 736 private <R> R runAction(Action<R> action, R errorResult, String method, 737 boolean reconnect, boolean onlyEstablishedConnection) { 738 synchronized (mStartLock) { 739 if (mServiceConnection == null) { 740 Log.w(TAG, method + " failed: not bound to TTS engine"); 741 return errorResult; 742 } 743 return mServiceConnection.runAction(action, errorResult, method, reconnect, 744 onlyEstablishedConnection); 745 } 746 } 747 748 private int initTts() { 749 // Step 1: Try connecting to the engine that was requested. 750 if (mRequestedEngine != null) { 751 if (mEnginesHelper.isEngineInstalled(mRequestedEngine)) { 752 if (connectToEngine(mRequestedEngine)) { 753 mCurrentEngine = mRequestedEngine; 754 return SUCCESS; 755 } else if (!mUseFallback) { 756 mCurrentEngine = null; 757 dispatchOnInit(ERROR); 758 return ERROR; 759 } 760 } else if (!mUseFallback) { 761 Log.i(TAG, "Requested engine not installed: " + mRequestedEngine); 762 mCurrentEngine = null; 763 dispatchOnInit(ERROR); 764 return ERROR; 765 } 766 } 767 768 // Step 2: Try connecting to the user's default engine. 769 final String defaultEngine = getDefaultEngine(); 770 if (defaultEngine != null && !defaultEngine.equals(mRequestedEngine)) { 771 if (connectToEngine(defaultEngine)) { 772 mCurrentEngine = defaultEngine; 773 return SUCCESS; 774 } 775 } 776 777 // Step 3: Try connecting to the highest ranked engine in the 778 // system. 779 final String highestRanked = mEnginesHelper.getHighestRankedEngineName(); 780 if (highestRanked != null && !highestRanked.equals(mRequestedEngine) && 781 !highestRanked.equals(defaultEngine)) { 782 if (connectToEngine(highestRanked)) { 783 mCurrentEngine = highestRanked; 784 return SUCCESS; 785 } 786 } 787 788 // NOTE: The API currently does not allow the caller to query whether 789 // they are actually connected to any engine. This might fail for various 790 // reasons like if the user disables all her TTS engines. 791 792 mCurrentEngine = null; 793 dispatchOnInit(ERROR); 794 return ERROR; 795 } 796 797 private boolean connectToEngine(String engine) { 798 Connection connection = new Connection(); 799 Intent intent = new Intent(Engine.INTENT_ACTION_TTS_SERVICE); 800 intent.setPackage(engine); 801 boolean bound = mContext.bindService(intent, connection, Context.BIND_AUTO_CREATE); 802 if (!bound) { 803 Log.e(TAG, "Failed to bind to " + engine); 804 return false; 805 } else { 806 Log.i(TAG, "Sucessfully bound to " + engine); 807 mConnectingServiceConnection = connection; 808 return true; 809 } 810 } 811 812 private void dispatchOnInit(int result) { 813 synchronized (mStartLock) { 814 if (mInitListener != null) { 815 mInitListener.onInit(result); 816 mInitListener = null; 817 } 818 } 819 } 820 821 private IBinder getCallerIdentity() { 822 return mServiceConnection.getCallerIdentity(); 823 } 824 825 /** 826 * Releases the resources used by the TextToSpeech engine. 827 * It is good practice for instance to call this method in the onDestroy() method of an Activity 828 * so the TextToSpeech engine can be cleanly stopped. 829 */ 830 public void shutdown() { 831 // Special case, we are asked to shutdown connection that did finalize its connection. 832 synchronized (mStartLock) { 833 if (mConnectingServiceConnection != null) { 834 mContext.unbindService(mConnectingServiceConnection); 835 mConnectingServiceConnection = null; 836 return; 837 } 838 } 839 840 // Post connection case 841 runActionNoReconnect(new Action<Void>() { 842 @Override 843 public Void run(ITextToSpeechService service) throws RemoteException { 844 service.setCallback(getCallerIdentity(), null); 845 service.stop(getCallerIdentity()); 846 mServiceConnection.disconnect(); 847 // Context#unbindService does not result in a call to 848 // ServiceConnection#onServiceDisconnected. As a result, the 849 // service ends up being destroyed (if there are no other open 850 // connections to it) but the process lives on and the 851 // ServiceConnection continues to refer to the destroyed service. 852 // 853 // This leads to tons of log spam about SynthThread being dead. 854 mServiceConnection = null; 855 mCurrentEngine = null; 856 return null; 857 } 858 }, null, "shutdown", false); 859 } 860 861 /** 862 * Adds a mapping between a string of text and a sound resource in a 863 * package. After a call to this method, subsequent calls to 864 * {@link #speak(String, int, HashMap)} will play the specified sound resource 865 * if it is available, or synthesize the text it is missing. 866 * 867 * @param text 868 * The string of text. Example: <code>"south_south_east"</code> 869 * 870 * @param packagename 871 * Pass the packagename of the application that contains the 872 * resource. If the resource is in your own application (this is 873 * the most common case), then put the packagename of your 874 * application here.<br/> 875 * Example: <b>"com.google.marvin.compass"</b><br/> 876 * The packagename can be found in the AndroidManifest.xml of 877 * your application. 878 * <p> 879 * <code><manifest xmlns:android="..." 880 * package="<b>com.google.marvin.compass</b>"></code> 881 * </p> 882 * 883 * @param resourceId 884 * Example: <code>R.raw.south_south_east</code> 885 * 886 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 887 */ 888 public int addSpeech(String text, String packagename, @RawRes int resourceId) { 889 synchronized (mStartLock) { 890 mUtterances.put(text, makeResourceUri(packagename, resourceId)); 891 return SUCCESS; 892 } 893 } 894 895 /** 896 * Adds a mapping between a CharSequence (may be spanned with TtsSpans) of text 897 * and a sound resource in a package. After a call to this method, subsequent calls to 898 * {@link #speak(String, int, HashMap)} will play the specified sound resource 899 * if it is available, or synthesize the text it is missing. 900 * 901 * @param text 902 * The string of text. Example: <code>"south_south_east"</code> 903 * 904 * @param packagename 905 * Pass the packagename of the application that contains the 906 * resource. If the resource is in your own application (this is 907 * the most common case), then put the packagename of your 908 * application here.<br/> 909 * Example: <b>"com.google.marvin.compass"</b><br/> 910 * The packagename can be found in the AndroidManifest.xml of 911 * your application. 912 * <p> 913 * <code><manifest xmlns:android="..." 914 * package="<b>com.google.marvin.compass</b>"></code> 915 * </p> 916 * 917 * @param resourceId 918 * Example: <code>R.raw.south_south_east</code> 919 * 920 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 921 */ 922 public int addSpeech(CharSequence text, String packagename, int resourceId) { 923 synchronized (mStartLock) { 924 mUtterances.put(text, makeResourceUri(packagename, resourceId)); 925 return SUCCESS; 926 } 927 } 928 929 /** 930 * Adds a mapping between a string of text and a sound file. Using this, it 931 * is possible to add custom pronounciations for a string of text. 932 * After a call to this method, subsequent calls to {@link #speak(String, int, HashMap)} 933 * will play the specified sound resource if it is available, or synthesize the text it is 934 * missing. 935 * 936 * @param text 937 * The string of text. Example: <code>"south_south_east"</code> 938 * @param filename 939 * The full path to the sound file (for example: 940 * "/sdcard/mysounds/hello.wav") 941 * 942 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 943 */ 944 public int addSpeech(String text, String filename) { 945 synchronized (mStartLock) { 946 mUtterances.put(text, Uri.parse(filename)); 947 return SUCCESS; 948 } 949 } 950 951 /** 952 * Adds a mapping between a CharSequence (may be spanned with TtsSpans and a sound file. 953 * Using this, it is possible to add custom pronounciations for a string of text. 954 * After a call to this method, subsequent calls to {@link #speak(String, int, HashMap)} 955 * will play the specified sound resource if it is available, or synthesize the text it is 956 * missing. 957 * 958 * @param text 959 * The string of text. Example: <code>"south_south_east"</code> 960 * @param file 961 * File object pointing to the sound file. 962 * 963 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 964 */ 965 public int addSpeech(CharSequence text, File file) { 966 synchronized (mStartLock) { 967 mUtterances.put(text, Uri.fromFile(file)); 968 return SUCCESS; 969 } 970 } 971 972 /** 973 * Adds a mapping between a string of text and a sound resource in a 974 * package. Use this to add custom earcons. 975 * 976 * @see #playEarcon(String, int, HashMap) 977 * 978 * @param earcon The name of the earcon. 979 * Example: <code>"[tick]"</code><br/> 980 * 981 * @param packagename 982 * the package name of the application that contains the 983 * resource. This can for instance be the package name of your own application. 984 * Example: <b>"com.google.marvin.compass"</b><br/> 985 * The package name can be found in the AndroidManifest.xml of 986 * the application containing the resource. 987 * <p> 988 * <code><manifest xmlns:android="..." 989 * package="<b>com.google.marvin.compass</b>"></code> 990 * </p> 991 * 992 * @param resourceId 993 * Example: <code>R.raw.tick_snd</code> 994 * 995 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 996 */ 997 public int addEarcon(String earcon, String packagename, @RawRes int resourceId) { 998 synchronized(mStartLock) { 999 mEarcons.put(earcon, makeResourceUri(packagename, resourceId)); 1000 return SUCCESS; 1001 } 1002 } 1003 1004 /** 1005 * Adds a mapping between a string of text and a sound file. 1006 * Use this to add custom earcons. 1007 * 1008 * @see #playEarcon(String, int, HashMap) 1009 * 1010 * @param earcon 1011 * The name of the earcon. 1012 * Example: <code>"[tick]"</code> 1013 * @param filename 1014 * The full path to the sound file (for example: 1015 * "/sdcard/mysounds/tick.wav") 1016 * 1017 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 1018 * 1019 * @deprecated As of API level 21, replaced by 1020 * {@link #addEarcon(String, File)}. 1021 */ 1022 @Deprecated 1023 public int addEarcon(String earcon, String filename) { 1024 synchronized(mStartLock) { 1025 mEarcons.put(earcon, Uri.parse(filename)); 1026 return SUCCESS; 1027 } 1028 } 1029 1030 /** 1031 * Adds a mapping between a string of text and a sound file. 1032 * Use this to add custom earcons. 1033 * 1034 * @see #playEarcon(String, int, HashMap) 1035 * 1036 * @param earcon 1037 * The name of the earcon. 1038 * Example: <code>"[tick]"</code> 1039 * @param file 1040 * File object pointing to the sound file. 1041 * 1042 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 1043 */ 1044 public int addEarcon(String earcon, File file) { 1045 synchronized(mStartLock) { 1046 mEarcons.put(earcon, Uri.fromFile(file)); 1047 return SUCCESS; 1048 } 1049 } 1050 1051 private Uri makeResourceUri(String packageName, int resourceId) { 1052 return new Uri.Builder() 1053 .scheme(ContentResolver.SCHEME_ANDROID_RESOURCE) 1054 .encodedAuthority(packageName) 1055 .appendEncodedPath(String.valueOf(resourceId)) 1056 .build(); 1057 } 1058 1059 /** 1060 * Speaks the text using the specified queuing strategy and speech parameters, the text may 1061 * be spanned with TtsSpans. 1062 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1063 * requests and then returns. The synthesis might not have finished (or even started!) at the 1064 * time when this method returns. In order to reliably detect errors during synthesis, 1065 * we recommend setting an utterance progress listener (see 1066 * {@link #setOnUtteranceProgressListener}) and using the 1067 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1068 * 1069 * @param text The string of text to be spoken. No longer than 1070 * {@link #getMaxSpeechInputLength()} characters. 1071 * @param queueMode The queuing strategy to use, {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1072 * @param params Parameters for the request. Can be null. 1073 * Supported parameter names: 1074 * {@link Engine#KEY_PARAM_STREAM}, 1075 * {@link Engine#KEY_PARAM_VOLUME}, 1076 * {@link Engine#KEY_PARAM_PAN}. 1077 * Engine specific parameters may be passed in but the parameter keys 1078 * must be prefixed by the name of the engine they are intended for. For example 1079 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1080 * engine named "com.svox.pico" if it is being used. 1081 * @param utteranceId An unique identifier for this request. 1082 * 1083 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the speak operation. 1084 */ 1085 public int speak(final CharSequence text, 1086 final int queueMode, 1087 final Bundle params, 1088 final String utteranceId) { 1089 return runAction(new Action<Integer>() { 1090 @Override 1091 public Integer run(ITextToSpeechService service) throws RemoteException { 1092 Uri utteranceUri = mUtterances.get(text); 1093 if (utteranceUri != null) { 1094 return service.playAudio(getCallerIdentity(), utteranceUri, queueMode, 1095 getParams(params), utteranceId); 1096 } else { 1097 return service.speak(getCallerIdentity(), text, queueMode, getParams(params), 1098 utteranceId); 1099 } 1100 } 1101 }, ERROR, "speak"); 1102 } 1103 1104 /** 1105 * Speaks the string using the specified queuing strategy and speech parameters. 1106 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1107 * requests and then returns. The synthesis might not have finished (or even started!) at the 1108 * time when this method returns. In order to reliably detect errors during synthesis, 1109 * we recommend setting an utterance progress listener (see 1110 * {@link #setOnUtteranceProgressListener}) and using the 1111 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1112 * 1113 * @param text The string of text to be spoken. No longer than 1114 * {@link #getMaxSpeechInputLength()} characters. 1115 * @param queueMode The queuing strategy to use, {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1116 * @param params Parameters for the request. Can be null. 1117 * Supported parameter names: 1118 * {@link Engine#KEY_PARAM_STREAM}, 1119 * {@link Engine#KEY_PARAM_UTTERANCE_ID}, 1120 * {@link Engine#KEY_PARAM_VOLUME}, 1121 * {@link Engine#KEY_PARAM_PAN}. 1122 * Engine specific parameters may be passed in but the parameter keys 1123 * must be prefixed by the name of the engine they are intended for. For example 1124 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1125 * engine named "com.svox.pico" if it is being used. 1126 * 1127 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the speak operation. 1128 * @deprecated As of API level 21, replaced by 1129 * {@link #speak(CharSequence, int, Bundle, String)}. 1130 */ 1131 @Deprecated 1132 public int speak(final String text, final int queueMode, final HashMap<String, String> params) { 1133 return speak(text, queueMode, convertParamsHashMaptoBundle(params), 1134 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1135 } 1136 1137 /** 1138 * Plays the earcon using the specified queueing mode and parameters. 1139 * The earcon must already have been added with {@link #addEarcon(String, String)} or 1140 * {@link #addEarcon(String, String, int)}. 1141 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1142 * requests and then returns. The synthesis might not have finished (or even started!) at the 1143 * time when this method returns. In order to reliably detect errors during synthesis, 1144 * we recommend setting an utterance progress listener (see 1145 * {@link #setOnUtteranceProgressListener}) and using the 1146 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1147 * 1148 * @param earcon The earcon that should be played 1149 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1150 * @param params Parameters for the request. Can be null. 1151 * Supported parameter names: 1152 * {@link Engine#KEY_PARAM_STREAM}, 1153 * Engine specific parameters may be passed in but the parameter keys 1154 * must be prefixed by the name of the engine they are intended for. For example 1155 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1156 * engine named "com.svox.pico" if it is being used. 1157 * 1158 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playEarcon operation. 1159 */ 1160 public int playEarcon(final String earcon, final int queueMode, 1161 final Bundle params, final String utteranceId) { 1162 return runAction(new Action<Integer>() { 1163 @Override 1164 public Integer run(ITextToSpeechService service) throws RemoteException { 1165 Uri earconUri = mEarcons.get(earcon); 1166 if (earconUri == null) { 1167 return ERROR; 1168 } 1169 return service.playAudio(getCallerIdentity(), earconUri, queueMode, 1170 getParams(params), utteranceId); 1171 } 1172 }, ERROR, "playEarcon"); 1173 } 1174 1175 /** 1176 * Plays the earcon using the specified queueing mode and parameters. 1177 * The earcon must already have been added with {@link #addEarcon(String, String)} or 1178 * {@link #addEarcon(String, String, int)}. 1179 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1180 * requests and then returns. The synthesis might not have finished (or even started!) at the 1181 * time when this method returns. In order to reliably detect errors during synthesis, 1182 * we recommend setting an utterance progress listener (see 1183 * {@link #setOnUtteranceProgressListener}) and using the 1184 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1185 * 1186 * @param earcon The earcon that should be played 1187 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1188 * @param params Parameters for the request. Can be null. 1189 * Supported parameter names: 1190 * {@link Engine#KEY_PARAM_STREAM}, 1191 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1192 * Engine specific parameters may be passed in but the parameter keys 1193 * must be prefixed by the name of the engine they are intended for. For example 1194 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1195 * engine named "com.svox.pico" if it is being used. 1196 * 1197 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playEarcon operation. 1198 * @deprecated As of API level 21, replaced by 1199 * {@link #playEarcon(String, int, Bundle, String)}. 1200 */ 1201 @Deprecated 1202 public int playEarcon(final String earcon, final int queueMode, 1203 final HashMap<String, String> params) { 1204 return playEarcon(earcon, queueMode, convertParamsHashMaptoBundle(params), 1205 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1206 } 1207 1208 /** 1209 * Plays silence for the specified amount of time using the specified 1210 * queue mode. 1211 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1212 * requests and then returns. The synthesis might not have finished (or even started!) at the 1213 * time when this method returns. In order to reliably detect errors during synthesis, 1214 * we recommend setting an utterance progress listener (see 1215 * {@link #setOnUtteranceProgressListener}) and using the 1216 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1217 * 1218 * @param durationInMs The duration of the silence. 1219 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1220 * @param utteranceId An unique identifier for this request. 1221 * 1222 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playSilentUtterance operation. 1223 */ 1224 public int playSilentUtterance(final long durationInMs, final int queueMode, 1225 final String utteranceId) { 1226 return runAction(new Action<Integer>() { 1227 @Override 1228 public Integer run(ITextToSpeechService service) throws RemoteException { 1229 return service.playSilence(getCallerIdentity(), durationInMs, 1230 queueMode, utteranceId); 1231 } 1232 }, ERROR, "playSilentUtterance"); 1233 } 1234 1235 /** 1236 * Plays silence for the specified amount of time using the specified 1237 * queue mode. 1238 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1239 * requests and then returns. The synthesis might not have finished (or even started!) at the 1240 * time when this method returns. In order to reliably detect errors during synthesis, 1241 * we recommend setting an utterance progress listener (see 1242 * {@link #setOnUtteranceProgressListener}) and using the 1243 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1244 * 1245 * @param durationInMs The duration of the silence. 1246 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1247 * @param params Parameters for the request. Can be null. 1248 * Supported parameter names: 1249 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1250 * Engine specific parameters may be passed in but the parameter keys 1251 * must be prefixed by the name of the engine they are intended for. For example 1252 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1253 * engine named "com.svox.pico" if it is being used. 1254 * 1255 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playSilence operation. 1256 * @deprecated As of API level 21, replaced by 1257 * {@link #playSilentUtterance(long, int, String)}. 1258 */ 1259 @Deprecated 1260 public int playSilence(final long durationInMs, final int queueMode, 1261 final HashMap<String, String> params) { 1262 return playSilentUtterance(durationInMs, queueMode, 1263 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1264 } 1265 1266 /** 1267 * Queries the engine for the set of features it supports for a given locale. 1268 * Features can either be framework defined, e.g. 1269 * {@link TextToSpeech.Engine#KEY_FEATURE_NETWORK_SYNTHESIS} or engine specific. 1270 * Engine specific keys must be prefixed by the name of the engine they 1271 * are intended for. These keys can be used as parameters to 1272 * {@link TextToSpeech#speak(String, int, java.util.HashMap)} and 1273 * {@link TextToSpeech#synthesizeToFile(String, java.util.HashMap, String)}. 1274 * 1275 * Features values are strings and their values must meet restrictions described in their 1276 * documentation. 1277 * 1278 * @param locale The locale to query features for. 1279 * @return Set instance. May return {@code null} on error. 1280 * @deprecated As of API level 21, please use voices. In order to query features of the voice, 1281 * call {@link #getVoices()} to retrieve the list of available voices and 1282 * {@link Voice#getFeatures()} to retrieve the set of features. 1283 */ 1284 @Deprecated 1285 public Set<String> getFeatures(final Locale locale) { 1286 return runAction(new Action<Set<String>>() { 1287 @Override 1288 public Set<String> run(ITextToSpeechService service) throws RemoteException { 1289 String[] features = null; 1290 try { 1291 features = service.getFeaturesForLanguage( 1292 locale.getISO3Language(), locale.getISO3Country(), locale.getVariant()); 1293 } catch(MissingResourceException e) { 1294 Log.w(TAG, "Couldn't retrieve 3 letter ISO 639-2/T language and/or ISO 3166 " + 1295 "country code for locale: " + locale, e); 1296 return null; 1297 } 1298 1299 if (features != null) { 1300 final Set<String> featureSet = new HashSet<String>(); 1301 Collections.addAll(featureSet, features); 1302 return featureSet; 1303 } 1304 return null; 1305 } 1306 }, null, "getFeatures"); 1307 } 1308 1309 /** 1310 * Checks whether the TTS engine is busy speaking. Note that a speech item is 1311 * considered complete once it's audio data has been sent to the audio mixer, or 1312 * written to a file. There might be a finite lag between this point, and when 1313 * the audio hardware completes playback. 1314 * 1315 * @return {@code true} if the TTS engine is speaking. 1316 */ 1317 public boolean isSpeaking() { 1318 return runAction(new Action<Boolean>() { 1319 @Override 1320 public Boolean run(ITextToSpeechService service) throws RemoteException { 1321 return service.isSpeaking(); 1322 } 1323 }, false, "isSpeaking"); 1324 } 1325 1326 /** 1327 * Interrupts the current utterance (whether played or rendered to file) and discards other 1328 * utterances in the queue. 1329 * 1330 * @return {@link #ERROR} or {@link #SUCCESS}. 1331 */ 1332 public int stop() { 1333 return runAction(new Action<Integer>() { 1334 @Override 1335 public Integer run(ITextToSpeechService service) throws RemoteException { 1336 return service.stop(getCallerIdentity()); 1337 } 1338 }, ERROR, "stop"); 1339 } 1340 1341 /** 1342 * Sets the speech rate. 1343 * 1344 * This has no effect on any pre-recorded speech. 1345 * 1346 * @param speechRate Speech rate. {@code 1.0} is the normal speech rate, 1347 * lower values slow down the speech ({@code 0.5} is half the normal speech rate), 1348 * greater values accelerate it ({@code 2.0} is twice the normal speech rate). 1349 * 1350 * @return {@link #ERROR} or {@link #SUCCESS}. 1351 */ 1352 public int setSpeechRate(float speechRate) { 1353 if (speechRate > 0.0f) { 1354 int intRate = (int)(speechRate * 100); 1355 if (intRate > 0) { 1356 synchronized (mStartLock) { 1357 mParams.putInt(Engine.KEY_PARAM_RATE, intRate); 1358 } 1359 return SUCCESS; 1360 } 1361 } 1362 return ERROR; 1363 } 1364 1365 /** 1366 * Sets the speech pitch for the TextToSpeech engine. 1367 * 1368 * This has no effect on any pre-recorded speech. 1369 * 1370 * @param pitch Speech pitch. {@code 1.0} is the normal pitch, 1371 * lower values lower the tone of the synthesized voice, 1372 * greater values increase it. 1373 * 1374 * @return {@link #ERROR} or {@link #SUCCESS}. 1375 */ 1376 public int setPitch(float pitch) { 1377 if (pitch > 0.0f) { 1378 int intPitch = (int)(pitch * 100); 1379 if (intPitch > 0) { 1380 synchronized (mStartLock) { 1381 mParams.putInt(Engine.KEY_PARAM_PITCH, intPitch); 1382 } 1383 return SUCCESS; 1384 } 1385 } 1386 return ERROR; 1387 } 1388 1389 /** 1390 * Sets the audio attributes to be used when speaking text or playing 1391 * back a file. 1392 * 1393 * @param audioAttributes Valid AudioAttributes instance. 1394 * 1395 * @return {@link #ERROR} or {@link #SUCCESS}. 1396 */ 1397 public int setAudioAttributes(AudioAttributes audioAttributes) { 1398 if (audioAttributes != null) { 1399 synchronized (mStartLock) { 1400 mParams.putParcelable(Engine.KEY_PARAM_AUDIO_ATTRIBUTES, 1401 audioAttributes); 1402 } 1403 return SUCCESS; 1404 } 1405 return ERROR; 1406 } 1407 1408 /** 1409 * @return the engine currently in use by this TextToSpeech instance. 1410 * @hide 1411 */ 1412 public String getCurrentEngine() { 1413 return mCurrentEngine; 1414 } 1415 1416 /** 1417 * Returns a Locale instance describing the language currently being used as the default 1418 * Text-to-speech language. 1419 * 1420 * The locale object returned by this method is NOT a valid one. It has identical form to the 1421 * one in {@link #getLanguage()}. Please refer to {@link #getLanguage()} for more information. 1422 * 1423 * @return language, country (if any) and variant (if any) used by the client stored in a 1424 * Locale instance, or {@code null} on error. 1425 * @deprecated As of API level 21, use <code>getDefaultVoice().getLocale()</code> ({@link 1426 * #getDefaultVoice()}) 1427 */ 1428 @Deprecated 1429 public Locale getDefaultLanguage() { 1430 return runAction(new Action<Locale>() { 1431 @Override 1432 public Locale run(ITextToSpeechService service) throws RemoteException { 1433 String[] defaultLanguage = service.getClientDefaultLanguage(); 1434 1435 return new Locale(defaultLanguage[0], defaultLanguage[1], defaultLanguage[2]); 1436 } 1437 }, null, "getDefaultLanguage"); 1438 } 1439 1440 /** 1441 * Sets the text-to-speech language. 1442 * The TTS engine will try to use the closest match to the specified 1443 * language as represented by the Locale, but there is no guarantee that the exact same Locale 1444 * will be used. Use {@link #isLanguageAvailable(Locale)} to check the level of support 1445 * before choosing the language to use for the next utterances. 1446 * 1447 * This method sets the current voice to the default one for the given Locale; 1448 * {@link #getVoice()} can be used to retrieve it. 1449 * 1450 * @param loc The locale describing the language to be used. 1451 * 1452 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1453 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1454 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1455 */ 1456 public int setLanguage(final Locale loc) { 1457 return runAction(new Action<Integer>() { 1458 @Override 1459 public Integer run(ITextToSpeechService service) throws RemoteException { 1460 if (loc == null) { 1461 return LANG_NOT_SUPPORTED; 1462 } 1463 String language = null, country = null; 1464 try { 1465 language = loc.getISO3Language(); 1466 } catch (MissingResourceException e) { 1467 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1468 return LANG_NOT_SUPPORTED; 1469 } 1470 1471 try { 1472 country = loc.getISO3Country(); 1473 } catch (MissingResourceException e) { 1474 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1475 return LANG_NOT_SUPPORTED; 1476 } 1477 1478 String variant = loc.getVariant(); 1479 1480 // As of API level 21, setLanguage is implemented using setVoice. 1481 // (which, in the default implementation, will call loadLanguage on the service 1482 // interface). 1483 1484 // Sanitize locale using isLanguageAvailable. 1485 int result = service.isLanguageAvailable( language, country, variant); 1486 if (result >= LANG_AVAILABLE){ 1487 if (result < LANG_COUNTRY_VAR_AVAILABLE) { 1488 variant = ""; 1489 if (result < LANG_COUNTRY_AVAILABLE) { 1490 country = ""; 1491 } 1492 } 1493 // Get the default voice for the locale. 1494 String voiceName = service.getDefaultVoiceNameFor(language, country, variant); 1495 if (TextUtils.isEmpty(voiceName)) { 1496 Log.w(TAG, "Couldn't find the default voice for " + language + "/" + 1497 country + "/" + variant); 1498 return LANG_NOT_SUPPORTED; 1499 } 1500 1501 // Load it. 1502 if (service.loadVoice(getCallerIdentity(), voiceName) == TextToSpeech.ERROR) { 1503 return LANG_NOT_SUPPORTED; 1504 } 1505 1506 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, voiceName); 1507 mParams.putString(Engine.KEY_PARAM_LANGUAGE, language); 1508 mParams.putString(Engine.KEY_PARAM_COUNTRY, country); 1509 mParams.putString(Engine.KEY_PARAM_VARIANT, variant); 1510 } 1511 return result; 1512 } 1513 }, LANG_NOT_SUPPORTED, "setLanguage"); 1514 } 1515 1516 /** 1517 * Returns a Locale instance describing the language currently being used for synthesis 1518 * requests sent to the TextToSpeech engine. 1519 * 1520 * In Android 4.2 and before (API <= 17) this function returns the language that is currently 1521 * being used by the TTS engine. That is the last language set by this or any other 1522 * client by a {@link TextToSpeech#setLanguage} call to the same engine. 1523 * 1524 * In Android versions after 4.2 this function returns the language that is currently being 1525 * used for the synthesis requests sent from this client. That is the last language set 1526 * by a {@link TextToSpeech#setLanguage} call on this instance. 1527 * 1528 * If a voice is set (by {@link #setVoice(Voice)}), getLanguage will return the language of 1529 * the currently set voice. 1530 * 1531 * Please note that the Locale object returned by this method is NOT a valid Locale object. Its 1532 * language field contains a three-letter ISO 639-2/T code (where a proper Locale would use 1533 * a two-letter ISO 639-1 code), and the country field contains a three-letter ISO 3166 country 1534 * code (where a proper Locale would use a two-letter ISO 3166-1 code). 1535 * 1536 * @return language, country (if any) and variant (if any) used by the client stored in a 1537 * Locale instance, or {@code null} on error. 1538 * 1539 * @deprecated As of API level 21, please use <code>getVoice().getLocale()</code> 1540 * ({@link #getVoice()}). 1541 */ 1542 @Deprecated 1543 public Locale getLanguage() { 1544 return runAction(new Action<Locale>() { 1545 @Override 1546 public Locale run(ITextToSpeechService service) { 1547 /* No service call, but we're accessing mParams, hence need for 1548 wrapping it as an Action instance */ 1549 String lang = mParams.getString(Engine.KEY_PARAM_LANGUAGE, ""); 1550 String country = mParams.getString(Engine.KEY_PARAM_COUNTRY, ""); 1551 String variant = mParams.getString(Engine.KEY_PARAM_VARIANT, ""); 1552 return new Locale(lang, country, variant); 1553 } 1554 }, null, "getLanguage"); 1555 } 1556 1557 /** 1558 * Query the engine about the set of available languages. 1559 */ 1560 public Set<Locale> getAvailableLanguages() { 1561 return runAction(new Action<Set<Locale>>() { 1562 @Override 1563 public Set<Locale> run(ITextToSpeechService service) throws RemoteException { 1564 List<Voice> voices = service.getVoices(); 1565 if (voices == null) { 1566 return new HashSet<Locale>(); 1567 } 1568 HashSet<Locale> locales = new HashSet<Locale>(); 1569 for (Voice voice : voices) { 1570 locales.add(voice.getLocale()); 1571 } 1572 return locales; 1573 } 1574 }, null, "getAvailableLanguages"); 1575 } 1576 1577 /** 1578 * Query the engine about the set of available voices. 1579 * 1580 * Each TTS Engine can expose multiple voices for each locale, each with a different set of 1581 * features. 1582 * 1583 * @see #setVoice(Voice) 1584 * @see Voice 1585 */ 1586 public Set<Voice> getVoices() { 1587 return runAction(new Action<Set<Voice>>() { 1588 @Override 1589 public Set<Voice> run(ITextToSpeechService service) throws RemoteException { 1590 List<Voice> voices = service.getVoices(); 1591 return (voices != null) ? new HashSet<Voice>(voices) : new HashSet<Voice>(); 1592 } 1593 }, null, "getVoices"); 1594 } 1595 1596 /** 1597 * Sets the text-to-speech voice. 1598 * 1599 * @param voice One of objects returned by {@link #getVoices()}. 1600 * 1601 * @return {@link #ERROR} or {@link #SUCCESS}. 1602 * 1603 * @see #getVoices 1604 * @see Voice 1605 */ 1606 public int setVoice(final Voice voice) { 1607 return runAction(new Action<Integer>() { 1608 @Override 1609 public Integer run(ITextToSpeechService service) throws RemoteException { 1610 int result = service.loadVoice(getCallerIdentity(), voice.getName()); 1611 if (result == SUCCESS) { 1612 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, voice.getName()); 1613 1614 // Set the language/country/variant, so #getLanguage will return the voice 1615 // locale when called. 1616 String language = ""; 1617 try { 1618 language = voice.getLocale().getISO3Language(); 1619 } catch (MissingResourceException e) { 1620 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + 1621 voice.getLocale(), e); 1622 } 1623 1624 String country = ""; 1625 try { 1626 country = voice.getLocale().getISO3Country(); 1627 } catch (MissingResourceException e) { 1628 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + 1629 voice.getLocale(), e); 1630 } 1631 mParams.putString(Engine.KEY_PARAM_LANGUAGE, language); 1632 mParams.putString(Engine.KEY_PARAM_COUNTRY, country); 1633 mParams.putString(Engine.KEY_PARAM_VARIANT, voice.getLocale().getVariant()); 1634 } 1635 return result; 1636 } 1637 }, LANG_NOT_SUPPORTED, "setVoice"); 1638 } 1639 1640 /** 1641 * Returns a Voice instance describing the voice currently being used for synthesis 1642 * requests sent to the TextToSpeech engine. 1643 * 1644 * @return Voice instance used by the client, or {@code null} if not set or on error. 1645 * 1646 * @see #getVoices 1647 * @see #setVoice 1648 * @see Voice 1649 */ 1650 public Voice getVoice() { 1651 return runAction(new Action<Voice>() { 1652 @Override 1653 public Voice run(ITextToSpeechService service) throws RemoteException { 1654 String voiceName = mParams.getString(Engine.KEY_PARAM_VOICE_NAME, ""); 1655 if (TextUtils.isEmpty(voiceName)) { 1656 return null; 1657 } 1658 List<Voice> voices = service.getVoices(); 1659 if (voices == null) { 1660 return null; 1661 } 1662 for (Voice voice : voices) { 1663 if (voice.getName().equals(voiceName)) { 1664 return voice; 1665 } 1666 } 1667 return null; 1668 } 1669 }, null, "getVoice"); 1670 } 1671 1672 /** 1673 * Returns a Voice instance that's the default voice for the default Text-to-speech language. 1674 * @return The default voice instance for the default language, or {@code null} if not set or 1675 * on error. 1676 */ 1677 public Voice getDefaultVoice() { 1678 return runAction(new Action<Voice>() { 1679 @Override 1680 public Voice run(ITextToSpeechService service) throws RemoteException { 1681 1682 String[] defaultLanguage = service.getClientDefaultLanguage(); 1683 1684 if (defaultLanguage == null || defaultLanguage.length == 0) { 1685 Log.e(TAG, "service.getClientDefaultLanguage() returned empty array"); 1686 return null; 1687 } 1688 String language = defaultLanguage[0]; 1689 String country = (defaultLanguage.length > 1) ? defaultLanguage[1] : ""; 1690 String variant = (defaultLanguage.length > 2) ? defaultLanguage[2] : ""; 1691 1692 // Sanitize the locale using isLanguageAvailable. 1693 int result = service.isLanguageAvailable(language, country, variant); 1694 if (result >= LANG_AVAILABLE){ 1695 if (result < LANG_COUNTRY_VAR_AVAILABLE) { 1696 variant = ""; 1697 if (result < LANG_COUNTRY_AVAILABLE) { 1698 country = ""; 1699 } 1700 } 1701 } else { 1702 // The default language is not supported. 1703 return null; 1704 } 1705 1706 // Get the default voice name 1707 String voiceName = service.getDefaultVoiceNameFor(language, country, variant); 1708 if (TextUtils.isEmpty(voiceName)) { 1709 return null; 1710 } 1711 1712 // Find it 1713 List<Voice> voices = service.getVoices(); 1714 if (voices == null) { 1715 return null; 1716 } 1717 for (Voice voice : voices) { 1718 if (voice.getName().equals(voiceName)) { 1719 return voice; 1720 } 1721 } 1722 return null; 1723 } 1724 }, null, "getDefaultVoice"); 1725 } 1726 1727 1728 1729 /** 1730 * Checks if the specified language as represented by the Locale is available and supported. 1731 * 1732 * @param loc The Locale describing the language to be used. 1733 * 1734 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1735 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1736 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1737 */ 1738 public int isLanguageAvailable(final Locale loc) { 1739 return runAction(new Action<Integer>() { 1740 @Override 1741 public Integer run(ITextToSpeechService service) throws RemoteException { 1742 String language = null, country = null; 1743 1744 try { 1745 language = loc.getISO3Language(); 1746 } catch (MissingResourceException e) { 1747 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1748 return LANG_NOT_SUPPORTED; 1749 } 1750 1751 try { 1752 country = loc.getISO3Country(); 1753 } catch (MissingResourceException e) { 1754 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1755 return LANG_NOT_SUPPORTED; 1756 } 1757 1758 return service.isLanguageAvailable(language, country, loc.getVariant()); 1759 } 1760 }, LANG_NOT_SUPPORTED, "isLanguageAvailable"); 1761 } 1762 1763 /** 1764 * Synthesizes the given text to a file using the specified parameters. 1765 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1766 * requests and then returns. The synthesis might not have finished (or even started!) at the 1767 * time when this method returns. In order to reliably detect errors during synthesis, 1768 * we recommend setting an utterance progress listener (see 1769 * {@link #setOnUtteranceProgressListener}). 1770 * 1771 * @param text The text that should be synthesized. No longer than 1772 * {@link #getMaxSpeechInputLength()} characters. 1773 * @param params Parameters for the request. Can be null. 1774 * Engine specific parameters may be passed in but the parameter keys 1775 * must be prefixed by the name of the engine they are intended for. For example 1776 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1777 * engine named "com.svox.pico" if it is being used. 1778 * @param file File to write the generated audio data to. 1779 * @param utteranceId An unique identifier for this request. 1780 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the synthesizeToFile operation. 1781 */ 1782 public int synthesizeToFile(final CharSequence text, final Bundle params, 1783 final File file, final String utteranceId) { 1784 return runAction(new Action<Integer>() { 1785 @Override 1786 public Integer run(ITextToSpeechService service) throws RemoteException { 1787 ParcelFileDescriptor fileDescriptor; 1788 int returnValue; 1789 try { 1790 if(file.exists() && !file.canWrite()) { 1791 Log.e(TAG, "Can't write to " + file); 1792 return ERROR; 1793 } 1794 fileDescriptor = ParcelFileDescriptor.open(file, 1795 ParcelFileDescriptor.MODE_WRITE_ONLY | 1796 ParcelFileDescriptor.MODE_CREATE | 1797 ParcelFileDescriptor.MODE_TRUNCATE); 1798 returnValue = service.synthesizeToFileDescriptor(getCallerIdentity(), text, 1799 fileDescriptor, getParams(params), utteranceId); 1800 fileDescriptor.close(); 1801 return returnValue; 1802 } catch (FileNotFoundException e) { 1803 Log.e(TAG, "Opening file " + file + " failed", e); 1804 return ERROR; 1805 } catch (IOException e) { 1806 Log.e(TAG, "Closing file " + file + " failed", e); 1807 return ERROR; 1808 } 1809 } 1810 }, ERROR, "synthesizeToFile"); 1811 } 1812 1813 /** 1814 * Synthesizes the given text to a file using the specified parameters. 1815 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1816 * requests and then returns. The synthesis might not have finished (or even started!) at the 1817 * time when this method returns. In order to reliably detect errors during synthesis, 1818 * we recommend setting an utterance progress listener (see 1819 * {@link #setOnUtteranceProgressListener}) and using the 1820 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1821 * 1822 * @param text The text that should be synthesized. No longer than 1823 * {@link #getMaxSpeechInputLength()} characters. 1824 * @param params Parameters for the request. Can be null. 1825 * Supported parameter names: 1826 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1827 * Engine specific parameters may be passed in but the parameter keys 1828 * must be prefixed by the name of the engine they are intended for. For example 1829 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1830 * engine named "com.svox.pico" if it is being used. 1831 * @param filename Absolute file filename to write the generated audio data to.It should be 1832 * something like "/sdcard/myappsounds/mysound.wav". 1833 * 1834 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the synthesizeToFile operation. 1835 * @deprecated As of API level 21, replaced by 1836 * {@link #synthesizeToFile(CharSequence, Bundle, File, String)}. 1837 */ 1838 @Deprecated 1839 public int synthesizeToFile(final String text, final HashMap<String, String> params, 1840 final String filename) { 1841 return synthesizeToFile(text, convertParamsHashMaptoBundle(params), 1842 new File(filename), params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1843 } 1844 1845 private Bundle convertParamsHashMaptoBundle(HashMap<String, String> params) { 1846 if (params != null && !params.isEmpty()) { 1847 Bundle bundle = new Bundle(); 1848 copyIntParam(bundle, params, Engine.KEY_PARAM_STREAM); 1849 copyIntParam(bundle, params, Engine.KEY_PARAM_SESSION_ID); 1850 copyStringParam(bundle, params, Engine.KEY_PARAM_UTTERANCE_ID); 1851 copyFloatParam(bundle, params, Engine.KEY_PARAM_VOLUME); 1852 copyFloatParam(bundle, params, Engine.KEY_PARAM_PAN); 1853 1854 // Copy feature strings defined by the framework. 1855 copyStringParam(bundle, params, Engine.KEY_FEATURE_NETWORK_SYNTHESIS); 1856 copyStringParam(bundle, params, Engine.KEY_FEATURE_EMBEDDED_SYNTHESIS); 1857 copyIntParam(bundle, params, Engine.KEY_FEATURE_NETWORK_TIMEOUT_MS); 1858 copyIntParam(bundle, params, Engine.KEY_FEATURE_NETWORK_RETRIES_COUNT); 1859 1860 // Copy over all parameters that start with the name of the 1861 // engine that we are currently connected to. The engine is 1862 // free to interpret them as it chooses. 1863 if (!TextUtils.isEmpty(mCurrentEngine)) { 1864 for (Map.Entry<String, String> entry : params.entrySet()) { 1865 final String key = entry.getKey(); 1866 if (key != null && key.startsWith(mCurrentEngine)) { 1867 bundle.putString(key, entry.getValue()); 1868 } 1869 } 1870 } 1871 1872 return bundle; 1873 } 1874 return null; 1875 } 1876 1877 private Bundle getParams(Bundle params) { 1878 if (params != null && !params.isEmpty()) { 1879 Bundle bundle = new Bundle(mParams); 1880 bundle.putAll(params); 1881 1882 verifyIntegerBundleParam(bundle, Engine.KEY_PARAM_STREAM); 1883 verifyIntegerBundleParam(bundle, Engine.KEY_PARAM_SESSION_ID); 1884 verifyStringBundleParam(bundle, Engine.KEY_PARAM_UTTERANCE_ID); 1885 verifyFloatBundleParam(bundle, Engine.KEY_PARAM_VOLUME); 1886 verifyFloatBundleParam(bundle, Engine.KEY_PARAM_PAN); 1887 1888 // Copy feature strings defined by the framework. 1889 verifyBooleanBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_SYNTHESIS); 1890 verifyBooleanBundleParam(bundle, Engine.KEY_FEATURE_EMBEDDED_SYNTHESIS); 1891 verifyIntegerBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_TIMEOUT_MS); 1892 verifyIntegerBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_RETRIES_COUNT); 1893 1894 return bundle; 1895 } else { 1896 return mParams; 1897 } 1898 } 1899 1900 private static boolean verifyIntegerBundleParam(Bundle bundle, String key) { 1901 if (bundle.containsKey(key)) { 1902 if (!(bundle.get(key) instanceof Integer || 1903 bundle.get(key) instanceof Long)) { 1904 bundle.remove(key); 1905 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1906 + " with invalid type. Should be an Integer or a Long"); 1907 return false; 1908 } 1909 } 1910 return true; 1911 } 1912 1913 private static boolean verifyStringBundleParam(Bundle bundle, String key) { 1914 if (bundle.containsKey(key)) { 1915 if (!(bundle.get(key) instanceof String)) { 1916 bundle.remove(key); 1917 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1918 + " with invalid type. Should be a String"); 1919 return false; 1920 } 1921 } 1922 return true; 1923 } 1924 1925 private static boolean verifyBooleanBundleParam(Bundle bundle, String key) { 1926 if (bundle.containsKey(key)) { 1927 if (!(bundle.get(key) instanceof Boolean || 1928 bundle.get(key) instanceof String)) { 1929 bundle.remove(key); 1930 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1931 + " with invalid type. Should be a Boolean or String"); 1932 return false; 1933 } 1934 } 1935 return true; 1936 } 1937 1938 1939 private static boolean verifyFloatBundleParam(Bundle bundle, String key) { 1940 if (bundle.containsKey(key)) { 1941 if (!(bundle.get(key) instanceof Float || 1942 bundle.get(key) instanceof Double)) { 1943 bundle.remove(key); 1944 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1945 + " with invalid type. Should be a Float or a Double"); 1946 return false; 1947 } 1948 } 1949 return true; 1950 } 1951 1952 private void copyStringParam(Bundle bundle, HashMap<String, String> params, String key) { 1953 String value = params.get(key); 1954 if (value != null) { 1955 bundle.putString(key, value); 1956 } 1957 } 1958 1959 private void copyIntParam(Bundle bundle, HashMap<String, String> params, String key) { 1960 String valueString = params.get(key); 1961 if (!TextUtils.isEmpty(valueString)) { 1962 try { 1963 int value = Integer.parseInt(valueString); 1964 bundle.putInt(key, value); 1965 } catch (NumberFormatException ex) { 1966 // don't set the value in the bundle 1967 } 1968 } 1969 } 1970 1971 private void copyFloatParam(Bundle bundle, HashMap<String, String> params, String key) { 1972 String valueString = params.get(key); 1973 if (!TextUtils.isEmpty(valueString)) { 1974 try { 1975 float value = Float.parseFloat(valueString); 1976 bundle.putFloat(key, value); 1977 } catch (NumberFormatException ex) { 1978 // don't set the value in the bundle 1979 } 1980 } 1981 } 1982 1983 /** 1984 * Sets the listener that will be notified when synthesis of an utterance completes. 1985 * 1986 * @param listener The listener to use. 1987 * 1988 * @return {@link #ERROR} or {@link #SUCCESS}. 1989 * 1990 * @deprecated Use {@link #setOnUtteranceProgressListener(UtteranceProgressListener)} 1991 * instead. 1992 */ 1993 @Deprecated 1994 public int setOnUtteranceCompletedListener(final OnUtteranceCompletedListener listener) { 1995 mUtteranceProgressListener = UtteranceProgressListener.from(listener); 1996 return TextToSpeech.SUCCESS; 1997 } 1998 1999 /** 2000 * Sets the listener that will be notified of various events related to the 2001 * synthesis of a given utterance. 2002 * 2003 * See {@link UtteranceProgressListener} and 2004 * {@link TextToSpeech.Engine#KEY_PARAM_UTTERANCE_ID}. 2005 * 2006 * @param listener the listener to use. 2007 * @return {@link #ERROR} or {@link #SUCCESS} 2008 */ 2009 public int setOnUtteranceProgressListener(UtteranceProgressListener listener) { 2010 mUtteranceProgressListener = listener; 2011 return TextToSpeech.SUCCESS; 2012 } 2013 2014 /** 2015 * Sets the TTS engine to use. 2016 * 2017 * @deprecated This doesn't inform callers when the TTS engine has been 2018 * initialized. {@link #TextToSpeech(Context, OnInitListener, String)} 2019 * can be used with the appropriate engine name. Also, there is no 2020 * guarantee that the engine specified will be loaded. If it isn't 2021 * installed or disabled, the user / system wide defaults will apply. 2022 * 2023 * @param enginePackageName The package name for the synthesis engine (e.g. "com.svox.pico") 2024 * 2025 * @return {@link #ERROR} or {@link #SUCCESS}. 2026 */ 2027 @Deprecated 2028 public int setEngineByPackageName(String enginePackageName) { 2029 mRequestedEngine = enginePackageName; 2030 return initTts(); 2031 } 2032 2033 /** 2034 * Gets the package name of the default speech synthesis engine. 2035 * 2036 * @return Package name of the TTS engine that the user has chosen 2037 * as their default. 2038 */ 2039 public String getDefaultEngine() { 2040 return mEnginesHelper.getDefaultEngine(); 2041 } 2042 2043 /** 2044 * Checks whether the user's settings should override settings requested 2045 * by the calling application. As of the Ice cream sandwich release, 2046 * user settings never forcibly override the app's settings. 2047 */ 2048 @Deprecated 2049 public boolean areDefaultsEnforced() { 2050 return false; 2051 } 2052 2053 /** 2054 * Gets a list of all installed TTS engines. 2055 * 2056 * @return A list of engine info objects. The list can be empty, but never {@code null}. 2057 */ 2058 public List<EngineInfo> getEngines() { 2059 return mEnginesHelper.getEngines(); 2060 } 2061 2062 private class Connection implements ServiceConnection { 2063 private ITextToSpeechService mService; 2064 2065 private SetupConnectionAsyncTask mOnSetupConnectionAsyncTask; 2066 2067 private boolean mEstablished; 2068 2069 private final ITextToSpeechCallback.Stub mCallback = new ITextToSpeechCallback.Stub() { 2070 public void onStop(String utteranceId, boolean isStarted) throws RemoteException { 2071 UtteranceProgressListener listener = mUtteranceProgressListener; 2072 if (listener != null) { 2073 listener.onStop(utteranceId, isStarted); 2074 } 2075 }; 2076 2077 @Override 2078 public void onSuccess(String utteranceId) { 2079 UtteranceProgressListener listener = mUtteranceProgressListener; 2080 if (listener != null) { 2081 listener.onDone(utteranceId); 2082 } 2083 } 2084 2085 @Override 2086 public void onError(String utteranceId, int errorCode) { 2087 UtteranceProgressListener listener = mUtteranceProgressListener; 2088 if (listener != null) { 2089 listener.onError(utteranceId); 2090 } 2091 } 2092 2093 @Override 2094 public void onStart(String utteranceId) { 2095 UtteranceProgressListener listener = mUtteranceProgressListener; 2096 if (listener != null) { 2097 listener.onStart(utteranceId); 2098 } 2099 } 2100 }; 2101 2102 private class SetupConnectionAsyncTask extends AsyncTask<Void, Void, Integer> { 2103 private final ComponentName mName; 2104 2105 public SetupConnectionAsyncTask(ComponentName name) { 2106 mName = name; 2107 } 2108 2109 @Override 2110 protected Integer doInBackground(Void... params) { 2111 synchronized(mStartLock) { 2112 if (isCancelled()) { 2113 return null; 2114 } 2115 2116 try { 2117 mService.setCallback(getCallerIdentity(), mCallback); 2118 2119 if (mParams.getString(Engine.KEY_PARAM_LANGUAGE) == null) { 2120 String[] defaultLanguage = mService.getClientDefaultLanguage(); 2121 mParams.putString(Engine.KEY_PARAM_LANGUAGE, defaultLanguage[0]); 2122 mParams.putString(Engine.KEY_PARAM_COUNTRY, defaultLanguage[1]); 2123 mParams.putString(Engine.KEY_PARAM_VARIANT, defaultLanguage[2]); 2124 2125 // Get the default voice for the locale. 2126 String defaultVoiceName = mService.getDefaultVoiceNameFor( 2127 defaultLanguage[0], defaultLanguage[1], defaultLanguage[2]); 2128 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, defaultVoiceName); 2129 } 2130 2131 Log.i(TAG, "Set up connection to " + mName); 2132 return SUCCESS; 2133 } catch (RemoteException re) { 2134 Log.e(TAG, "Error connecting to service, setCallback() failed"); 2135 return ERROR; 2136 } 2137 } 2138 } 2139 2140 @Override 2141 protected void onPostExecute(Integer result) { 2142 synchronized(mStartLock) { 2143 if (mOnSetupConnectionAsyncTask == this) { 2144 mOnSetupConnectionAsyncTask = null; 2145 } 2146 mEstablished = true; 2147 dispatchOnInit(result); 2148 } 2149 } 2150 } 2151 2152 @Override 2153 public void onServiceConnected(ComponentName name, IBinder service) { 2154 synchronized(mStartLock) { 2155 mConnectingServiceConnection = null; 2156 2157 Log.i(TAG, "Connected to " + name); 2158 2159 if (mOnSetupConnectionAsyncTask != null) { 2160 mOnSetupConnectionAsyncTask.cancel(false); 2161 } 2162 2163 mService = ITextToSpeechService.Stub.asInterface(service); 2164 mServiceConnection = Connection.this; 2165 2166 mEstablished = false; 2167 mOnSetupConnectionAsyncTask = new SetupConnectionAsyncTask(name); 2168 mOnSetupConnectionAsyncTask.execute(); 2169 } 2170 } 2171 2172 public IBinder getCallerIdentity() { 2173 return mCallback; 2174 } 2175 2176 /** 2177 * Clear connection related fields and cancel mOnServiceConnectedAsyncTask if set. 2178 * 2179 * @return true if we cancel mOnSetupConnectionAsyncTask in progress. 2180 */ 2181 private boolean clearServiceConnection() { 2182 synchronized(mStartLock) { 2183 boolean result = false; 2184 if (mOnSetupConnectionAsyncTask != null) { 2185 result = mOnSetupConnectionAsyncTask.cancel(false); 2186 mOnSetupConnectionAsyncTask = null; 2187 } 2188 2189 mService = null; 2190 // If this is the active connection, clear it 2191 if (mServiceConnection == this) { 2192 mServiceConnection = null; 2193 } 2194 return result; 2195 } 2196 } 2197 2198 @Override 2199 public void onServiceDisconnected(ComponentName name) { 2200 Log.i(TAG, "Asked to disconnect from " + name); 2201 if (clearServiceConnection()) { 2202 /* We need to protect against a rare case where engine 2203 * dies just after successful connection - and we process onServiceDisconnected 2204 * before OnServiceConnectedAsyncTask.onPostExecute. onServiceDisconnected cancels 2205 * OnServiceConnectedAsyncTask.onPostExecute and we don't call dispatchOnInit 2206 * with ERROR as argument. 2207 */ 2208 dispatchOnInit(ERROR); 2209 } 2210 } 2211 2212 public void disconnect() { 2213 mContext.unbindService(this); 2214 clearServiceConnection(); 2215 } 2216 2217 public boolean isEstablished() { 2218 return mService != null && mEstablished; 2219 } 2220 2221 public <R> R runAction(Action<R> action, R errorResult, String method, 2222 boolean reconnect, boolean onlyEstablishedConnection) { 2223 synchronized (mStartLock) { 2224 try { 2225 if (mService == null) { 2226 Log.w(TAG, method + " failed: not connected to TTS engine"); 2227 return errorResult; 2228 } 2229 if (onlyEstablishedConnection && !isEstablished()) { 2230 Log.w(TAG, method + " failed: TTS engine connection not fully set up"); 2231 return errorResult; 2232 } 2233 return action.run(mService); 2234 } catch (RemoteException ex) { 2235 Log.e(TAG, method + " failed", ex); 2236 if (reconnect) { 2237 disconnect(); 2238 initTts(); 2239 } 2240 return errorResult; 2241 } 2242 } 2243 } 2244 } 2245 2246 private interface Action<R> { 2247 R run(ITextToSpeechService service) throws RemoteException; 2248 } 2249 2250 /** 2251 * Information about an installed text-to-speech engine. 2252 * 2253 * @see TextToSpeech#getEngines 2254 */ 2255 public static class EngineInfo { 2256 /** 2257 * Engine package name.. 2258 */ 2259 public String name; 2260 /** 2261 * Localized label for the engine. 2262 */ 2263 public String label; 2264 /** 2265 * Icon for the engine. 2266 */ 2267 public int icon; 2268 /** 2269 * Whether this engine is a part of the system 2270 * image. 2271 * 2272 * @hide 2273 */ 2274 public boolean system; 2275 /** 2276 * The priority the engine declares for the the intent filter 2277 * {@code android.intent.action.TTS_SERVICE} 2278 * 2279 * @hide 2280 */ 2281 public int priority; 2282 2283 @Override 2284 public String toString() { 2285 return "EngineInfo{name=" + name + "}"; 2286 } 2287 2288 } 2289 2290 /** 2291 * Limit of length of input string passed to speak and synthesizeToFile. 2292 * 2293 * @see #speak 2294 * @see #synthesizeToFile 2295 */ 2296 public static int getMaxSpeechInputLength() { 2297 return 4000; 2298 } 2299} 2300