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 */ 16 package android.speech.tts; 17 18 import android.annotation.IntDef; 19 import android.annotation.Nullable; 20 import android.annotation.RawRes; 21 import android.annotation.SdkConstant; 22 import android.annotation.SdkConstant.SdkConstantType; 23 import android.content.ComponentName; 24 import android.content.ContentResolver; 25 import android.content.Context; 26 import android.content.Intent; 27 import android.content.ServiceConnection; 28 import android.media.AudioAttributes; 29 import android.media.AudioManager; 30 import android.net.Uri; 31 import android.os.AsyncTask; 32 import android.os.Bundle; 33 import android.os.IBinder; 34 import android.os.ParcelFileDescriptor; 35 import android.os.RemoteException; 36 import android.provider.Settings; 37 import android.text.TextUtils; 38 import android.util.Log; 39 40 import java.io.File; 41 import java.io.FileNotFoundException; 42 import java.io.IOException; 43 import java.lang.annotation.Retention; 44 import java.lang.annotation.RetentionPolicy; 45 import java.util.Collections; 46 import java.util.HashMap; 47 import java.util.HashSet; 48 import java.util.List; 49 import java.util.Locale; 50 import java.util.Map; 51 import java.util.MissingResourceException; 52 import java.util.Set; 53 54 /** 55 * 56 * Synthesizes speech from text for immediate playback or to create a sound file. 57 * <p>A TextToSpeech instance can only be used to synthesize text once it has completed its 58 * initialization. Implement the {@link TextToSpeech.OnInitListener} to be 59 * notified of the completion of the initialization.<br> 60 * When you are done using the TextToSpeech instance, call the {@link #shutdown()} method 61 * to release the native resources used by the TextToSpeech engine. 62 */ 63 public class TextToSpeech { 64 65 private static final String TAG = "TextToSpeech"; 66 67 /** 68 * Denotes a successful operation. 69 */ 70 public static final int SUCCESS = 0; 71 /** 72 * Denotes a generic operation failure. 73 */ 74 public static final int ERROR = -1; 75 76 /** 77 * Denotes a stop requested by a client. It's used only on the service side of the API, 78 * client should never expect to see this result code. 79 */ 80 public static final int STOPPED = -2; 81 82 /** @hide */ 83 @IntDef({ERROR_SYNTHESIS, ERROR_SERVICE, ERROR_OUTPUT, ERROR_NETWORK, ERROR_NETWORK_TIMEOUT, 84 ERROR_INVALID_REQUEST, ERROR_NOT_INSTALLED_YET}) 85 @Retention(RetentionPolicy.SOURCE) 86 public @interface Error {} 87 88 /** 89 * Denotes a failure of a TTS engine to synthesize the given input. 90 */ 91 public static final int ERROR_SYNTHESIS = -3; 92 93 /** 94 * Denotes a failure of a TTS service. 95 */ 96 public static final int ERROR_SERVICE = -4; 97 98 /** 99 * Denotes a failure related to the output (audio device or a file). 100 */ 101 public static final int ERROR_OUTPUT = -5; 102 103 /** 104 * Denotes a failure caused by a network connectivity problems. 105 */ 106 public static final int ERROR_NETWORK = -6; 107 108 /** 109 * Denotes a failure caused by network timeout. 110 */ 111 public static final int ERROR_NETWORK_TIMEOUT = -7; 112 113 /** 114 * Denotes a failure caused by an invalid request. 115 */ 116 public static final int ERROR_INVALID_REQUEST = -8; 117 118 /** 119 * Denotes a failure caused by an unfinished download of the voice data. 120 * @see Engine#KEY_FEATURE_NOT_INSTALLED 121 */ 122 public static final int ERROR_NOT_INSTALLED_YET = -9; 123 124 /** 125 * Queue mode where all entries in the playback queue (media to be played 126 * and text to be synthesized) are dropped and replaced by the new entry. 127 * Queues are flushed with respect to a given calling app. Entries in the queue 128 * from other callees are not discarded. 129 */ 130 public static final int QUEUE_FLUSH = 0; 131 /** 132 * Queue mode where the new entry is added at the end of the playback queue. 133 */ 134 public static final int QUEUE_ADD = 1; 135 /** 136 * Queue mode where the entire playback queue is purged. This is different 137 * from {@link #QUEUE_FLUSH} in that all entries are purged, not just entries 138 * from a given caller. 139 * 140 * @hide 141 */ 142 static final int QUEUE_DESTROY = 2; 143 144 /** 145 * Denotes the language is available exactly as specified by the locale. 146 */ 147 public static final int LANG_COUNTRY_VAR_AVAILABLE = 2; 148 149 /** 150 * Denotes the language is available for the language and country specified 151 * by the locale, but not the variant. 152 */ 153 public static final int LANG_COUNTRY_AVAILABLE = 1; 154 155 /** 156 * Denotes the language is available for the language by the locale, 157 * but not the country and variant. 158 */ 159 public static final int LANG_AVAILABLE = 0; 160 161 /** 162 * Denotes the language data is missing. 163 */ 164 public static final int LANG_MISSING_DATA = -1; 165 166 /** 167 * Denotes the language is not supported. 168 */ 169 public static final int LANG_NOT_SUPPORTED = -2; 170 171 /** 172 * Broadcast Action: The TextToSpeech synthesizer has completed processing 173 * of all the text in the speech queue. 174 * 175 * Note that this notifies callers when the <b>engine</b> has finished has 176 * processing text data. Audio playback might not have completed (or even started) 177 * at this point. If you wish to be notified when this happens, see 178 * {@link OnUtteranceCompletedListener}. 179 */ 180 @SdkConstant(SdkConstantType.BROADCAST_INTENT_ACTION) 181 public static final String ACTION_TTS_QUEUE_PROCESSING_COMPLETED = 182 "android.speech.tts.TTS_QUEUE_PROCESSING_COMPLETED"; 183 184 /** 185 * Interface definition of a callback to be invoked indicating the completion of the 186 * TextToSpeech engine initialization. 187 */ 188 public interface OnInitListener { 189 /** 190 * Called to signal the completion of the TextToSpeech engine initialization. 191 * 192 * @param status {@link TextToSpeech#SUCCESS} or {@link TextToSpeech#ERROR}. 193 */ 194 void onInit(int status); 195 } 196 197 /** 198 * Listener that will be called when the TTS service has 199 * completed synthesizing an utterance. This is only called if the utterance 200 * has an utterance ID (see {@link TextToSpeech.Engine#KEY_PARAM_UTTERANCE_ID}). 201 * 202 * @deprecated Use {@link UtteranceProgressListener} instead. 203 */ 204 @Deprecated 205 public interface OnUtteranceCompletedListener { 206 /** 207 * Called when an utterance has been synthesized. 208 * 209 * @param utteranceId the identifier of the utterance. 210 */ 211 void onUtteranceCompleted(String utteranceId); 212 } 213 214 /** 215 * Constants and parameter names for controlling text-to-speech. These include: 216 * 217 * <ul> 218 * <li> 219 * Intents to ask engine to install data or check its data and 220 * extras for a TTS engine's check data activity. 221 * </li> 222 * <li> 223 * Keys for the parameters passed with speak commands, e.g. 224 * {@link Engine#KEY_PARAM_UTTERANCE_ID}, {@link Engine#KEY_PARAM_STREAM}. 225 * </li> 226 * <li> 227 * A list of feature strings that engines might support, e.g 228 * {@link Engine#KEY_FEATURE_NETWORK_SYNTHESIS}. These values may be passed in to 229 * {@link TextToSpeech#speak} and {@link TextToSpeech#synthesizeToFile} to modify 230 * engine behaviour. The engine can be queried for the set of features it supports 231 * through {@link TextToSpeech#getFeatures(java.util.Locale)}. 232 * </li> 233 * </ul> 234 */ 235 public class Engine { 236 237 /** 238 * Default speech rate. 239 * @hide 240 */ 241 public static final int DEFAULT_RATE = 100; 242 243 /** 244 * Default pitch. 245 * @hide 246 */ 247 public static final int DEFAULT_PITCH = 100; 248 249 /** 250 * Default volume. 251 * @hide 252 */ 253 public static final float DEFAULT_VOLUME = 1.0f; 254 255 /** 256 * Default pan (centered). 257 * @hide 258 */ 259 public static final float DEFAULT_PAN = 0.0f; 260 261 /** 262 * Default value for {@link Settings.Secure#TTS_USE_DEFAULTS}. 263 * @hide 264 */ 265 public static final int USE_DEFAULTS = 0; // false 266 267 /** 268 * Package name of the default TTS engine. 269 * 270 * @hide 271 * @deprecated No longer in use, the default engine is determined by 272 * the sort order defined in {@link TtsEngines}. Note that 273 * this doesn't "break" anything because there is no guarantee that 274 * the engine specified below is installed on a given build, let 275 * alone be the default. 276 */ 277 @Deprecated 278 public static final String DEFAULT_ENGINE = "com.svox.pico"; 279 280 /** 281 * Default audio stream used when playing synthesized speech. 282 */ 283 public static final int DEFAULT_STREAM = AudioManager.STREAM_MUSIC; 284 285 /** 286 * Indicates success when checking the installation status of the resources used by the 287 * TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 288 */ 289 public static final int CHECK_VOICE_DATA_PASS = 1; 290 291 /** 292 * Indicates failure when checking the installation status of the resources used by the 293 * TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 294 */ 295 public static final int CHECK_VOICE_DATA_FAIL = 0; 296 297 /** 298 * Indicates erroneous data when checking the installation status of the resources used by 299 * 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_BAD_DATA = -1; 305 306 /** 307 * Indicates missing resources when checking the installation status of the resources used 308 * 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_DATA = -2; 314 315 /** 316 * Indicates missing storage volume when checking the installation status of the resources 317 * used by the TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 318 * 319 * @deprecated Use CHECK_VOICE_DATA_FAIL instead. 320 */ 321 @Deprecated 322 public static final int CHECK_VOICE_DATA_MISSING_VOLUME = -3; 323 324 /** 325 * Intent for starting a TTS service. Services that handle this intent must 326 * extend {@link TextToSpeechService}. Normal applications should not use this intent 327 * directly, instead they should talk to the TTS service using the the methods in this 328 * class. 329 */ 330 @SdkConstant(SdkConstantType.SERVICE_ACTION) 331 public static final String INTENT_ACTION_TTS_SERVICE = 332 "android.intent.action.TTS_SERVICE"; 333 334 /** 335 * Name under which a text to speech engine publishes information about itself. 336 * This meta-data should reference an XML resource containing a 337 * <code><{@link android.R.styleable#TextToSpeechEngine tts-engine}></code> 338 * tag. 339 */ 340 public static final String SERVICE_META_DATA = "android.speech.tts"; 341 342 // intents to ask engine to install data or check its data 343 /** 344 * Activity Action: Triggers the platform TextToSpeech engine to 345 * start the activity that installs the resource files on the device 346 * that are required for TTS to be operational. Since the installation 347 * of the data can be interrupted or declined by the user, the application 348 * shouldn't expect successful installation upon return from that intent, 349 * and if need be, should check installation status with 350 * {@link #ACTION_CHECK_TTS_DATA}. 351 */ 352 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 353 public static final String ACTION_INSTALL_TTS_DATA = 354 "android.speech.tts.engine.INSTALL_TTS_DATA"; 355 356 /** 357 * Broadcast Action: broadcast to signal the change in the list of available 358 * languages or/and their features. 359 */ 360 @SdkConstant(SdkConstantType.BROADCAST_INTENT_ACTION) 361 public static final String ACTION_TTS_DATA_INSTALLED = 362 "android.speech.tts.engine.TTS_DATA_INSTALLED"; 363 364 /** 365 * Activity Action: Starts the activity from the platform TextToSpeech 366 * engine to verify the proper installation and availability of the 367 * resource files on the system. Upon completion, the activity will 368 * return one of the following codes: 369 * {@link #CHECK_VOICE_DATA_PASS}, 370 * {@link #CHECK_VOICE_DATA_FAIL}, 371 * <p> Moreover, the data received in the activity result will contain the following 372 * fields: 373 * <ul> 374 * <li>{@link #EXTRA_AVAILABLE_VOICES} which contains an ArrayList<String> of all the 375 * available voices. The format of each voice is: lang-COUNTRY-variant where COUNTRY and 376 * variant are optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE").</li> 377 * <li>{@link #EXTRA_UNAVAILABLE_VOICES} which contains an ArrayList<String> of all the 378 * unavailable voices (ones that user can install). The format of each voice is: 379 * lang-COUNTRY-variant where COUNTRY and variant are optional (ie, "eng" or 380 * "eng-USA" or "eng-USA-FEMALE").</li> 381 * </ul> 382 */ 383 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 384 public static final String ACTION_CHECK_TTS_DATA = 385 "android.speech.tts.engine.CHECK_TTS_DATA"; 386 387 /** 388 * Activity intent for getting some sample text to use for demonstrating TTS. Specific 389 * locale have to be requested by passing following extra parameters: 390 * <ul> 391 * <li>language</li> 392 * <li>country</li> 393 * <li>variant</li> 394 * </ul> 395 * 396 * Upon completion, the activity result may contain the following fields: 397 * <ul> 398 * <li>{@link #EXTRA_SAMPLE_TEXT} which contains an String with sample text.</li> 399 * </ul> 400 */ 401 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 402 public static final String ACTION_GET_SAMPLE_TEXT = 403 "android.speech.tts.engine.GET_SAMPLE_TEXT"; 404 405 /** 406 * Extra information received with the {@link #ACTION_GET_SAMPLE_TEXT} intent result where 407 * the TextToSpeech engine returns an String with sample text for requested voice 408 */ 409 public static final String EXTRA_SAMPLE_TEXT = "sampleText"; 410 411 412 // extras for a TTS engine's check data activity 413 /** 414 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 415 * the TextToSpeech engine returns an ArrayList<String> of all the available voices. 416 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 417 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 418 */ 419 public static final String EXTRA_AVAILABLE_VOICES = "availableVoices"; 420 421 /** 422 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 423 * the TextToSpeech engine returns an ArrayList<String> of all the unavailable voices. 424 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 425 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 426 */ 427 public static final String EXTRA_UNAVAILABLE_VOICES = "unavailableVoices"; 428 429 /** 430 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 431 * the TextToSpeech engine specifies the path to its resources. 432 * 433 * It may be used by language packages to find out where to put their data. 434 * 435 * @deprecated TTS engine implementation detail, this information has no use for 436 * text-to-speech API client. 437 */ 438 @Deprecated 439 public static final String EXTRA_VOICE_DATA_ROOT_DIRECTORY = "dataRoot"; 440 441 /** 442 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 443 * the TextToSpeech engine specifies the file names of its resources under the 444 * resource path. 445 * 446 * @deprecated TTS engine implementation detail, this information has no use for 447 * text-to-speech API client. 448 */ 449 @Deprecated 450 public static final String EXTRA_VOICE_DATA_FILES = "dataFiles"; 451 452 /** 453 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 454 * the TextToSpeech engine specifies the locale associated with each resource file. 455 * 456 * @deprecated TTS engine implementation detail, this information has no use for 457 * text-to-speech API client. 458 */ 459 @Deprecated 460 public static final String EXTRA_VOICE_DATA_FILES_INFO = "dataFilesInfo"; 461 462 /** 463 * Extra information sent with the {@link #ACTION_CHECK_TTS_DATA} intent where the 464 * caller indicates to the TextToSpeech engine which specific sets of voice data to 465 * check for by sending an ArrayList<String> of the voices that are of interest. 466 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 467 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 468 * 469 * @deprecated Redundant functionality, checking for existence of specific sets of voice 470 * data can be done on client side. 471 */ 472 @Deprecated 473 public static final String EXTRA_CHECK_VOICE_DATA_FOR = "checkVoiceDataFor"; 474 475 // extras for a TTS engine's data installation 476 /** 477 * Extra information received with the {@link #ACTION_TTS_DATA_INSTALLED} intent result. 478 * It indicates whether the data files for the synthesis engine were successfully 479 * installed. The installation was initiated with the {@link #ACTION_INSTALL_TTS_DATA} 480 * intent. The possible values for this extra are 481 * {@link TextToSpeech#SUCCESS} and {@link TextToSpeech#ERROR}. 482 * 483 * @deprecated No longer in use. If client ise interested in information about what 484 * changed, is should send ACTION_CHECK_TTS_DATA intent to discover available voices. 485 */ 486 @Deprecated 487 public static final String EXTRA_TTS_DATA_INSTALLED = "dataInstalled"; 488 489 // keys for the parameters passed with speak commands. Hidden keys are used internally 490 // to maintain engine state for each TextToSpeech instance. 491 /** 492 * @hide 493 */ 494 public static final String KEY_PARAM_RATE = "rate"; 495 496 /** 497 * @hide 498 */ 499 public static final String KEY_PARAM_VOICE_NAME = "voiceName"; 500 501 /** 502 * @hide 503 */ 504 public static final String KEY_PARAM_LANGUAGE = "language"; 505 506 /** 507 * @hide 508 */ 509 public static final String KEY_PARAM_COUNTRY = "country"; 510 511 /** 512 * @hide 513 */ 514 public static final String KEY_PARAM_VARIANT = "variant"; 515 516 /** 517 * @hide 518 */ 519 public static final String KEY_PARAM_ENGINE = "engine"; 520 521 /** 522 * @hide 523 */ 524 public static final String KEY_PARAM_PITCH = "pitch"; 525 526 /** 527 * Parameter key to specify the audio stream type to be used when speaking text 528 * or playing back a file. The value should be one of the STREAM_ constants 529 * defined in {@link AudioManager}. 530 * 531 * @see TextToSpeech#speak(String, int, HashMap) 532 * @see TextToSpeech#playEarcon(String, int, HashMap) 533 */ 534 public static final String KEY_PARAM_STREAM = "streamType"; 535 536 /** 537 * Parameter key to specify the audio attributes to be used when 538 * speaking text or playing back a file. The value should be set 539 * using {@link TextToSpeech#setAudioAttributes(AudioAttributes)}. 540 * 541 * @see TextToSpeech#speak(String, int, HashMap) 542 * @see TextToSpeech#playEarcon(String, int, HashMap) 543 * @hide 544 */ 545 public static final String KEY_PARAM_AUDIO_ATTRIBUTES = "audioAttributes"; 546 547 /** 548 * Parameter key to identify an utterance in the 549 * {@link TextToSpeech.OnUtteranceCompletedListener} after text has been 550 * spoken, a file has been played back or a silence duration has elapsed. 551 * 552 * @see TextToSpeech#speak(String, int, HashMap) 553 * @see TextToSpeech#playEarcon(String, int, HashMap) 554 * @see TextToSpeech#synthesizeToFile(String, HashMap, String) 555 */ 556 public static final String KEY_PARAM_UTTERANCE_ID = "utteranceId"; 557 558 /** 559 * Parameter key to specify the speech volume relative to the current stream type 560 * volume used when speaking text. Volume is specified as a float ranging from 0 to 1 561 * where 0 is silence, and 1 is the maximum volume (the default behavior). 562 * 563 * @see TextToSpeech#speak(String, int, HashMap) 564 * @see TextToSpeech#playEarcon(String, int, HashMap) 565 */ 566 public static final String KEY_PARAM_VOLUME = "volume"; 567 568 /** 569 * Parameter key to specify how the speech is panned from left to right when speaking text. 570 * Pan is specified as a float ranging from -1 to +1 where -1 maps to a hard-left pan, 571 * 0 to center (the default behavior), and +1 to hard-right. 572 * 573 * @see TextToSpeech#speak(String, int, HashMap) 574 * @see TextToSpeech#playEarcon(String, int, HashMap) 575 */ 576 public static final String KEY_PARAM_PAN = "pan"; 577 578 /** 579 * Feature key for network synthesis. See {@link TextToSpeech#getFeatures(Locale)} 580 * for a description of how feature keys work. If set (and supported by the engine 581 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must 582 * use network based synthesis. 583 * 584 * @see TextToSpeech#speak(String, int, java.util.HashMap) 585 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 586 * @see TextToSpeech#getFeatures(java.util.Locale) 587 * 588 * @deprecated Starting from API level 21, to select network synthesis, call 589 * {@link TextToSpeech#getVoices()}, find a suitable network voice 590 * ({@link Voice#isNetworkConnectionRequired()}) and pass it 591 * to {@link TextToSpeech#setVoice(Voice)}. 592 */ 593 @Deprecated 594 public static final String KEY_FEATURE_NETWORK_SYNTHESIS = "networkTts"; 595 596 /** 597 * Feature key for embedded synthesis. See {@link TextToSpeech#getFeatures(Locale)} 598 * for a description of how feature keys work. If set and supported by the engine 599 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must synthesize 600 * text on-device (without making network requests). 601 * 602 * @see TextToSpeech#speak(String, int, java.util.HashMap) 603 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 604 * @see TextToSpeech#getFeatures(java.util.Locale) 605 606 * @deprecated Starting from API level 21, to select embedded synthesis, call 607 * ({@link TextToSpeech#getVoices()}, find a suitable embedded voice 608 * ({@link Voice#isNetworkConnectionRequired()}) and pass it 609 * to {@link TextToSpeech#setVoice(Voice)}). 610 */ 611 @Deprecated 612 public static final String KEY_FEATURE_EMBEDDED_SYNTHESIS = "embeddedTts"; 613 614 /** 615 * Parameter key to specify an audio session identifier (obtained from 616 * {@link AudioManager#generateAudioSessionId()}) that will be used by the request audio 617 * output. It can be used to associate one of the {@link android.media.audiofx.AudioEffect} 618 * objects with the synthesis (or earcon) output. 619 * 620 * @see TextToSpeech#speak(String, int, HashMap) 621 * @see TextToSpeech#playEarcon(String, int, HashMap) 622 */ 623 public static final String KEY_PARAM_SESSION_ID = "sessionId"; 624 625 /** 626 * Feature key that indicates that the voice may need to download additional data to be fully 627 * functional. The download will be triggered by calling 628 * {@link TextToSpeech#setVoice(Voice)} or {@link TextToSpeech#setLanguage(Locale)}. 629 * Until download is complete, each synthesis request will either report 630 * {@link TextToSpeech#ERROR_NOT_INSTALLED_YET} error, or use a different voice to synthesize 631 * the request. This feature should NOT be used as a key of a request parameter. 632 * 633 * @see TextToSpeech#getFeatures(java.util.Locale) 634 * @see Voice#getFeatures() 635 */ 636 public static final String KEY_FEATURE_NOT_INSTALLED = "notInstalled"; 637 638 /** 639 * Feature key that indicate that a network timeout can be set for the request. If set and 640 * supported as per {@link TextToSpeech#getFeatures(Locale)} or {@link Voice#getFeatures()}, 641 * it can be used as request parameter to set the maximum allowed time for a single 642 * request attempt, in milliseconds, before synthesis fails. When used as a key of 643 * a request parameter, its value should be a string with an integer value. 644 * 645 * @see TextToSpeech#getFeatures(java.util.Locale) 646 * @see Voice#getFeatures() 647 */ 648 public static final String KEY_FEATURE_NETWORK_TIMEOUT_MS = "networkTimeoutMs"; 649 650 /** 651 * Feature key that indicates that network request retries count can be set for the request. 652 * If set and supported as per {@link TextToSpeech#getFeatures(Locale)} or 653 * {@link Voice#getFeatures()}, it can be used as a request parameter to set the 654 * number of network request retries that are attempted in case of failure. When used as 655 * a key of a request parameter, its value should be a string with an integer value. 656 * 657 * @see TextToSpeech#getFeatures(java.util.Locale) 658 * @see Voice#getFeatures() 659 */ 660 public static final String KEY_FEATURE_NETWORK_RETRIES_COUNT = "networkRetriesCount"; 661 } 662 663 private final Context mContext; 664 private Connection mConnectingServiceConnection; 665 private Connection mServiceConnection; 666 private OnInitListener mInitListener; 667 // Written from an unspecified application thread, read from 668 // a binder thread. 669 @Nullable private volatile UtteranceProgressListener mUtteranceProgressListener; 670 private final Object mStartLock = new Object(); 671 672 private String mRequestedEngine; 673 // Whether to initialize this TTS object with the default engine, 674 // if the requested engine is not available. Valid only if mRequestedEngine 675 // is not null. Used only for testing, though potentially useful API wise 676 // too. 677 private final boolean mUseFallback; 678 private final Map<String, Uri> mEarcons; 679 private final Map<CharSequence, Uri> mUtterances; 680 private final Bundle mParams = new Bundle(); 681 private final TtsEngines mEnginesHelper; 682 private volatile String mCurrentEngine = null; 683 684 /** 685 * The constructor for the TextToSpeech class, using the default TTS engine. 686 * This will also initialize the associated TextToSpeech engine if it isn't already running. 687 * 688 * @param context 689 * The context this instance is running in. 690 * @param listener 691 * The {@link TextToSpeech.OnInitListener} that will be called when the 692 * TextToSpeech engine has initialized. In a case of a failure the listener 693 * may be called immediately, before TextToSpeech instance is fully constructed. 694 */ 695 public TextToSpeech(Context context, OnInitListener listener) { 696 this(context, listener, null); 697 } 698 699 /** 700 * The constructor for the TextToSpeech class, using the given TTS engine. 701 * This will also initialize the associated TextToSpeech engine if it isn't already running. 702 * 703 * @param context 704 * The context this instance is running in. 705 * @param listener 706 * The {@link TextToSpeech.OnInitListener} that will be called when the 707 * TextToSpeech engine has initialized. In a case of a failure the listener 708 * may be called immediately, before TextToSpeech instance is fully constructed. 709 * @param engine Package name of the TTS engine to use. 710 */ 711 public TextToSpeech(Context context, OnInitListener listener, String engine) { 712 this(context, listener, engine, null, true); 713 } 714 715 /** 716 * Used by the framework to instantiate TextToSpeech objects with a supplied 717 * package name, instead of using {@link android.content.Context#getPackageName()} 718 * 719 * @hide 720 */ 721 public TextToSpeech(Context context, OnInitListener listener, String engine, 722 String packageName, boolean useFallback) { 723 mContext = context; 724 mInitListener = listener; 725 mRequestedEngine = engine; 726 mUseFallback = useFallback; 727 728 mEarcons = new HashMap<String, Uri>(); 729 mUtterances = new HashMap<CharSequence, Uri>(); 730 mUtteranceProgressListener = null; 731 732 mEnginesHelper = new TtsEngines(mContext); 733 initTts(); 734 } 735 736 private <R> R runActionNoReconnect(Action<R> action, R errorResult, String method, 737 boolean onlyEstablishedConnection) { 738 return runAction(action, errorResult, method, false, onlyEstablishedConnection); 739 } 740 741 private <R> R runAction(Action<R> action, R errorResult, String method) { 742 return runAction(action, errorResult, method, true, true); 743 } 744 745 private <R> R runAction(Action<R> action, R errorResult, String method, 746 boolean reconnect, boolean onlyEstablishedConnection) { 747 synchronized (mStartLock) { 748 if (mServiceConnection == null) { 749 Log.w(TAG, method + " failed: not bound to TTS engine"); 750 return errorResult; 751 } 752 return mServiceConnection.runAction(action, errorResult, method, reconnect, 753 onlyEstablishedConnection); 754 } 755 } 756 757 private int initTts() { 758 // Step 1: Try connecting to the engine that was requested. 759 if (mRequestedEngine != null) { 760 if (mEnginesHelper.isEngineInstalled(mRequestedEngine)) { 761 if (connectToEngine(mRequestedEngine)) { 762 mCurrentEngine = mRequestedEngine; 763 return SUCCESS; 764 } else if (!mUseFallback) { 765 mCurrentEngine = null; 766 dispatchOnInit(ERROR); 767 return ERROR; 768 } 769 } else if (!mUseFallback) { 770 Log.i(TAG, "Requested engine not installed: " + mRequestedEngine); 771 mCurrentEngine = null; 772 dispatchOnInit(ERROR); 773 return ERROR; 774 } 775 } 776 777 // Step 2: Try connecting to the user's default engine. 778 final String defaultEngine = getDefaultEngine(); 779 if (defaultEngine != null && !defaultEngine.equals(mRequestedEngine)) { 780 if (connectToEngine(defaultEngine)) { 781 mCurrentEngine = defaultEngine; 782 return SUCCESS; 783 } 784 } 785 786 // Step 3: Try connecting to the highest ranked engine in the 787 // system. 788 final String highestRanked = mEnginesHelper.getHighestRankedEngineName(); 789 if (highestRanked != null && !highestRanked.equals(mRequestedEngine) && 790 !highestRanked.equals(defaultEngine)) { 791 if (connectToEngine(highestRanked)) { 792 mCurrentEngine = highestRanked; 793 return SUCCESS; 794 } 795 } 796 797 // NOTE: The API currently does not allow the caller to query whether 798 // they are actually connected to any engine. This might fail for various 799 // reasons like if the user disables all her TTS engines. 800 801 mCurrentEngine = null; 802 dispatchOnInit(ERROR); 803 return ERROR; 804 } 805 806 private boolean connectToEngine(String engine) { 807 Connection connection = new Connection(); 808 Intent intent = new Intent(Engine.INTENT_ACTION_TTS_SERVICE); 809 intent.setPackage(engine); 810 boolean bound = mContext.bindService(intent, connection, Context.BIND_AUTO_CREATE); 811 if (!bound) { 812 Log.e(TAG, "Failed to bind to " + engine); 813 return false; 814 } else { 815 Log.i(TAG, "Sucessfully bound to " + engine); 816 mConnectingServiceConnection = connection; 817 return true; 818 } 819 } 820 821 private void dispatchOnInit(int result) { 822 synchronized (mStartLock) { 823 if (mInitListener != null) { 824 mInitListener.onInit(result); 825 mInitListener = null; 826 } 827 } 828 } 829 830 private IBinder getCallerIdentity() { 831 return mServiceConnection.getCallerIdentity(); 832 } 833 834 /** 835 * Releases the resources used by the TextToSpeech engine. 836 * It is good practice for instance to call this method in the onDestroy() method of an Activity 837 * so the TextToSpeech engine can be cleanly stopped. 838 */ 839 public void shutdown() { 840 // Special case, we are asked to shutdown connection that did finalize its connection. 841 synchronized (mStartLock) { 842 if (mConnectingServiceConnection != null) { 843 mContext.unbindService(mConnectingServiceConnection); 844 mConnectingServiceConnection = null; 845 return; 846 } 847 } 848 849 // Post connection case 850 runActionNoReconnect(new Action<Void>() { 851 @Override 852 public Void run(ITextToSpeechService service) throws RemoteException { 853 service.setCallback(getCallerIdentity(), null); 854 service.stop(getCallerIdentity()); 855 mServiceConnection.disconnect(); 856 // Context#unbindService does not result in a call to 857 // ServiceConnection#onServiceDisconnected. As a result, the 858 // service ends up being destroyed (if there are no other open 859 // connections to it) but the process lives on and the 860 // ServiceConnection continues to refer to the destroyed service. 861 // 862 // This leads to tons of log spam about SynthThread being dead. 863 mServiceConnection = null; 864 mCurrentEngine = null; 865 return null; 866 } 867 }, null, "shutdown", false); 868 } 869 870 /** 871 * Adds a mapping between a string of text and a sound resource in a 872 * package. After a call to this method, subsequent calls to 873 * {@link #speak(String, int, HashMap)} will play the specified sound resource 874 * if it is available, or synthesize the text it is missing. 875 * 876 * @param text 877 * The string of text. Example: <code>"south_south_east"</code> 878 * 879 * @param packagename 880 * Pass the packagename of the application that contains the 881 * resource. If the resource is in your own application (this is 882 * the most common case), then put the packagename of your 883 * application here.<br/> 884 * Example: <b>"com.google.marvin.compass"</b><br/> 885 * The packagename can be found in the AndroidManifest.xml of 886 * your application. 887 * <p> 888 * <code><manifest xmlns:android="..." 889 * package="<b>com.google.marvin.compass</b>"></code> 890 * </p> 891 * 892 * @param resourceId 893 * Example: <code>R.raw.south_south_east</code> 894 * 895 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 896 */ 897 public int addSpeech(String text, String packagename, @RawRes int resourceId) { 898 synchronized (mStartLock) { 899 mUtterances.put(text, makeResourceUri(packagename, resourceId)); 900 return SUCCESS; 901 } 902 } 903 904 /** 905 * Adds a mapping between a CharSequence (may be spanned with TtsSpans) of text 906 * and a sound resource in a package. After a call to this method, subsequent calls to 907 * {@link #speak(String, int, HashMap)} will play the specified sound resource 908 * if it is available, or synthesize the text it is missing. 909 * 910 * @param text 911 * The string of text. Example: <code>"south_south_east"</code> 912 * 913 * @param packagename 914 * Pass the packagename of the application that contains the 915 * resource. If the resource is in your own application (this is 916 * the most common case), then put the packagename of your 917 * application here.<br/> 918 * Example: <b>"com.google.marvin.compass"</b><br/> 919 * The packagename can be found in the AndroidManifest.xml of 920 * your application. 921 * <p> 922 * <code><manifest xmlns:android="..." 923 * package="<b>com.google.marvin.compass</b>"></code> 924 * </p> 925 * 926 * @param resourceId 927 * Example: <code>R.raw.south_south_east</code> 928 * 929 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 930 */ 931 public int addSpeech(CharSequence text, String packagename, @RawRes int resourceId) { 932 synchronized (mStartLock) { 933 mUtterances.put(text, makeResourceUri(packagename, resourceId)); 934 return SUCCESS; 935 } 936 } 937 938 /** 939 * Adds a mapping between a string of text and a sound file. Using this, it 940 * is possible to add custom pronounciations for a string of text. 941 * After a call to this method, subsequent calls to {@link #speak(String, int, HashMap)} 942 * will play the specified sound resource if it is available, or synthesize the text it is 943 * missing. 944 * 945 * @param text 946 * The string of text. Example: <code>"south_south_east"</code> 947 * @param filename 948 * The full path to the sound file (for example: 949 * "/sdcard/mysounds/hello.wav") 950 * 951 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 952 */ 953 public int addSpeech(String text, String filename) { 954 synchronized (mStartLock) { 955 mUtterances.put(text, Uri.parse(filename)); 956 return SUCCESS; 957 } 958 } 959 960 /** 961 * Adds a mapping between a CharSequence (may be spanned with TtsSpans and a sound file. 962 * Using this, it is possible to add custom pronounciations for a string of text. 963 * After a call to this method, subsequent calls to {@link #speak(String, int, HashMap)} 964 * will play the specified sound resource if it is available, or synthesize the text it is 965 * missing. 966 * 967 * @param text 968 * The string of text. Example: <code>"south_south_east"</code> 969 * @param file 970 * File object pointing to the sound file. 971 * 972 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 973 */ 974 public int addSpeech(CharSequence text, File file) { 975 synchronized (mStartLock) { 976 mUtterances.put(text, Uri.fromFile(file)); 977 return SUCCESS; 978 } 979 } 980 981 /** 982 * Adds a mapping between a string of text and a sound resource in a 983 * package. Use this to add custom earcons. 984 * 985 * @see #playEarcon(String, int, HashMap) 986 * 987 * @param earcon The name of the earcon. 988 * Example: <code>"[tick]"</code><br/> 989 * 990 * @param packagename 991 * the package name of the application that contains the 992 * resource. This can for instance be the package name of your own application. 993 * Example: <b>"com.google.marvin.compass"</b><br/> 994 * The package name can be found in the AndroidManifest.xml of 995 * the application containing the resource. 996 * <p> 997 * <code><manifest xmlns:android="..." 998 * package="<b>com.google.marvin.compass</b>"></code> 999 * </p> 1000 * 1001 * @param resourceId 1002 * Example: <code>R.raw.tick_snd</code> 1003 * 1004 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 1005 */ 1006 public int addEarcon(String earcon, String packagename, @RawRes int resourceId) { 1007 synchronized(mStartLock) { 1008 mEarcons.put(earcon, makeResourceUri(packagename, resourceId)); 1009 return SUCCESS; 1010 } 1011 } 1012 1013 /** 1014 * Adds a mapping between a string of text and a sound file. 1015 * Use this to add custom earcons. 1016 * 1017 * @see #playEarcon(String, int, HashMap) 1018 * 1019 * @param earcon 1020 * The name of the earcon. 1021 * Example: <code>"[tick]"</code> 1022 * @param filename 1023 * The full path to the sound file (for example: 1024 * "/sdcard/mysounds/tick.wav") 1025 * 1026 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 1027 * 1028 * @deprecated As of API level 21, replaced by 1029 * {@link #addEarcon(String, File)}. 1030 */ 1031 @Deprecated 1032 public int addEarcon(String earcon, String filename) { 1033 synchronized(mStartLock) { 1034 mEarcons.put(earcon, Uri.parse(filename)); 1035 return SUCCESS; 1036 } 1037 } 1038 1039 /** 1040 * Adds a mapping between a string of text and a sound file. 1041 * Use this to add custom earcons. 1042 * 1043 * @see #playEarcon(String, int, HashMap) 1044 * 1045 * @param earcon 1046 * The name of the earcon. 1047 * Example: <code>"[tick]"</code> 1048 * @param file 1049 * File object pointing to the sound file. 1050 * 1051 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 1052 */ 1053 public int addEarcon(String earcon, File file) { 1054 synchronized(mStartLock) { 1055 mEarcons.put(earcon, Uri.fromFile(file)); 1056 return SUCCESS; 1057 } 1058 } 1059 1060 private Uri makeResourceUri(String packageName, int resourceId) { 1061 return new Uri.Builder() 1062 .scheme(ContentResolver.SCHEME_ANDROID_RESOURCE) 1063 .encodedAuthority(packageName) 1064 .appendEncodedPath(String.valueOf(resourceId)) 1065 .build(); 1066 } 1067 1068 /** 1069 * Speaks the text using the specified queuing strategy and speech parameters, the text may 1070 * be spanned with TtsSpans. 1071 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1072 * requests and then returns. The synthesis might not have finished (or even started!) at the 1073 * time when this method returns. In order to reliably detect errors during synthesis, 1074 * we recommend setting an utterance progress listener (see 1075 * {@link #setOnUtteranceProgressListener}) and using the 1076 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1077 * 1078 * @param text The string of text to be spoken. No longer than 1079 * {@link #getMaxSpeechInputLength()} characters. 1080 * @param queueMode The queuing strategy to use, {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1081 * @param params Parameters for the request. Can be null. 1082 * Supported parameter names: 1083 * {@link Engine#KEY_PARAM_STREAM}, 1084 * {@link Engine#KEY_PARAM_VOLUME}, 1085 * {@link Engine#KEY_PARAM_PAN}. 1086 * Engine specific parameters may be passed in but the parameter keys 1087 * must be prefixed by the name of the engine they are intended for. For example 1088 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1089 * engine named "com.svox.pico" if it is being used. 1090 * @param utteranceId An unique identifier for this request. 1091 * 1092 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the speak operation. 1093 */ 1094 public int speak(final CharSequence text, 1095 final int queueMode, 1096 final Bundle params, 1097 final String utteranceId) { 1098 return runAction(new Action<Integer>() { 1099 @Override 1100 public Integer run(ITextToSpeechService service) throws RemoteException { 1101 Uri utteranceUri = mUtterances.get(text); 1102 if (utteranceUri != null) { 1103 return service.playAudio(getCallerIdentity(), utteranceUri, queueMode, 1104 getParams(params), utteranceId); 1105 } else { 1106 return service.speak(getCallerIdentity(), text, queueMode, getParams(params), 1107 utteranceId); 1108 } 1109 } 1110 }, ERROR, "speak"); 1111 } 1112 1113 /** 1114 * Speaks the string using the specified queuing strategy and speech parameters. 1115 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1116 * requests and then returns. The synthesis might not have finished (or even started!) at the 1117 * time when this method returns. In order to reliably detect errors during synthesis, 1118 * we recommend setting an utterance progress listener (see 1119 * {@link #setOnUtteranceProgressListener}) and using the 1120 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1121 * 1122 * @param text The string of text to be spoken. No longer than 1123 * {@link #getMaxSpeechInputLength()} characters. 1124 * @param queueMode The queuing strategy to use, {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1125 * @param params Parameters for the request. Can be null. 1126 * Supported parameter names: 1127 * {@link Engine#KEY_PARAM_STREAM}, 1128 * {@link Engine#KEY_PARAM_UTTERANCE_ID}, 1129 * {@link Engine#KEY_PARAM_VOLUME}, 1130 * {@link Engine#KEY_PARAM_PAN}. 1131 * Engine specific parameters may be passed in but the parameter keys 1132 * must be prefixed by the name of the engine they are intended for. For example 1133 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1134 * engine named "com.svox.pico" if it is being used. 1135 * 1136 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the speak operation. 1137 * @deprecated As of API level 21, replaced by 1138 * {@link #speak(CharSequence, int, Bundle, String)}. 1139 */ 1140 @Deprecated 1141 public int speak(final String text, final int queueMode, final HashMap<String, String> params) { 1142 return speak(text, queueMode, convertParamsHashMaptoBundle(params), 1143 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1144 } 1145 1146 /** 1147 * Plays the earcon using the specified queueing mode and parameters. 1148 * The earcon must already have been added with {@link #addEarcon(String, String)} or 1149 * {@link #addEarcon(String, String, int)}. 1150 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1151 * requests and then returns. The synthesis might not have finished (or even started!) at the 1152 * time when this method returns. In order to reliably detect errors during synthesis, 1153 * we recommend setting an utterance progress listener (see 1154 * {@link #setOnUtteranceProgressListener}) and using the 1155 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1156 * 1157 * @param earcon The earcon that should be played 1158 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1159 * @param params Parameters for the request. Can be null. 1160 * Supported parameter names: 1161 * {@link Engine#KEY_PARAM_STREAM}, 1162 * Engine specific parameters may be passed in but the parameter keys 1163 * must be prefixed by the name of the engine they are intended for. For example 1164 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1165 * engine named "com.svox.pico" if it is being used. 1166 * 1167 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playEarcon operation. 1168 */ 1169 public int playEarcon(final String earcon, final int queueMode, 1170 final Bundle params, final String utteranceId) { 1171 return runAction(new Action<Integer>() { 1172 @Override 1173 public Integer run(ITextToSpeechService service) throws RemoteException { 1174 Uri earconUri = mEarcons.get(earcon); 1175 if (earconUri == null) { 1176 return ERROR; 1177 } 1178 return service.playAudio(getCallerIdentity(), earconUri, queueMode, 1179 getParams(params), utteranceId); 1180 } 1181 }, ERROR, "playEarcon"); 1182 } 1183 1184 /** 1185 * Plays the earcon using the specified queueing mode and parameters. 1186 * The earcon must already have been added with {@link #addEarcon(String, String)} or 1187 * {@link #addEarcon(String, String, int)}. 1188 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1189 * requests and then returns. The synthesis might not have finished (or even started!) at the 1190 * time when this method returns. In order to reliably detect errors during synthesis, 1191 * we recommend setting an utterance progress listener (see 1192 * {@link #setOnUtteranceProgressListener}) and using the 1193 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1194 * 1195 * @param earcon The earcon that should be played 1196 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1197 * @param params Parameters for the request. Can be null. 1198 * Supported parameter names: 1199 * {@link Engine#KEY_PARAM_STREAM}, 1200 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1201 * Engine specific parameters may be passed in but the parameter keys 1202 * must be prefixed by the name of the engine they are intended for. For example 1203 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1204 * engine named "com.svox.pico" if it is being used. 1205 * 1206 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playEarcon operation. 1207 * @deprecated As of API level 21, replaced by 1208 * {@link #playEarcon(String, int, Bundle, String)}. 1209 */ 1210 @Deprecated 1211 public int playEarcon(final String earcon, final int queueMode, 1212 final HashMap<String, String> params) { 1213 return playEarcon(earcon, queueMode, convertParamsHashMaptoBundle(params), 1214 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1215 } 1216 1217 /** 1218 * Plays silence for the specified amount of time using the specified 1219 * queue mode. 1220 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1221 * requests and then returns. The synthesis might not have finished (or even started!) at the 1222 * time when this method returns. In order to reliably detect errors during synthesis, 1223 * we recommend setting an utterance progress listener (see 1224 * {@link #setOnUtteranceProgressListener}) and using the 1225 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1226 * 1227 * @param durationInMs The duration of the silence. 1228 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1229 * @param utteranceId An unique identifier for this request. 1230 * 1231 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playSilentUtterance operation. 1232 */ 1233 public int playSilentUtterance(final long durationInMs, final int queueMode, 1234 final String utteranceId) { 1235 return runAction(new Action<Integer>() { 1236 @Override 1237 public Integer run(ITextToSpeechService service) throws RemoteException { 1238 return service.playSilence(getCallerIdentity(), durationInMs, 1239 queueMode, utteranceId); 1240 } 1241 }, ERROR, "playSilentUtterance"); 1242 } 1243 1244 /** 1245 * Plays silence for the specified amount of time using the specified 1246 * queue mode. 1247 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1248 * requests and then returns. The synthesis might not have finished (or even started!) at the 1249 * time when this method returns. In order to reliably detect errors during synthesis, 1250 * we recommend setting an utterance progress listener (see 1251 * {@link #setOnUtteranceProgressListener}) and using the 1252 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1253 * 1254 * @param durationInMs The duration of the silence. 1255 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1256 * @param params Parameters for the request. Can be null. 1257 * Supported parameter names: 1258 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1259 * Engine specific parameters may be passed in but the parameter keys 1260 * must be prefixed by the name of the engine they are intended for. For example 1261 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1262 * engine named "com.svox.pico" if it is being used. 1263 * 1264 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playSilence operation. 1265 * @deprecated As of API level 21, replaced by 1266 * {@link #playSilentUtterance(long, int, String)}. 1267 */ 1268 @Deprecated 1269 public int playSilence(final long durationInMs, final int queueMode, 1270 final HashMap<String, String> params) { 1271 return playSilentUtterance(durationInMs, queueMode, 1272 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1273 } 1274 1275 /** 1276 * Queries the engine for the set of features it supports for a given locale. 1277 * Features can either be framework defined, e.g. 1278 * {@link TextToSpeech.Engine#KEY_FEATURE_NETWORK_SYNTHESIS} or engine specific. 1279 * Engine specific keys must be prefixed by the name of the engine they 1280 * are intended for. These keys can be used as parameters to 1281 * {@link TextToSpeech#speak(String, int, java.util.HashMap)} and 1282 * {@link TextToSpeech#synthesizeToFile(String, java.util.HashMap, String)}. 1283 * 1284 * Features values are strings and their values must meet restrictions described in their 1285 * documentation. 1286 * 1287 * @param locale The locale to query features for. 1288 * @return Set instance. May return {@code null} on error. 1289 * @deprecated As of API level 21, please use voices. In order to query features of the voice, 1290 * call {@link #getVoices()} to retrieve the list of available voices and 1291 * {@link Voice#getFeatures()} to retrieve the set of features. 1292 */ 1293 @Deprecated 1294 public Set<String> getFeatures(final Locale locale) { 1295 return runAction(new Action<Set<String>>() { 1296 @Override 1297 public Set<String> run(ITextToSpeechService service) throws RemoteException { 1298 String[] features = null; 1299 try { 1300 features = service.getFeaturesForLanguage( 1301 locale.getISO3Language(), locale.getISO3Country(), locale.getVariant()); 1302 } catch(MissingResourceException e) { 1303 Log.w(TAG, "Couldn't retrieve 3 letter ISO 639-2/T language and/or ISO 3166 " + 1304 "country code for locale: " + locale, e); 1305 return null; 1306 } 1307 1308 if (features != null) { 1309 final Set<String> featureSet = new HashSet<String>(); 1310 Collections.addAll(featureSet, features); 1311 return featureSet; 1312 } 1313 return null; 1314 } 1315 }, null, "getFeatures"); 1316 } 1317 1318 /** 1319 * Checks whether the TTS engine is busy speaking. Note that a speech item is 1320 * considered complete once it's audio data has been sent to the audio mixer, or 1321 * written to a file. There might be a finite lag between this point, and when 1322 * the audio hardware completes playback. 1323 * 1324 * @return {@code true} if the TTS engine is speaking. 1325 */ 1326 public boolean isSpeaking() { 1327 return runAction(new Action<Boolean>() { 1328 @Override 1329 public Boolean run(ITextToSpeechService service) throws RemoteException { 1330 return service.isSpeaking(); 1331 } 1332 }, false, "isSpeaking"); 1333 } 1334 1335 /** 1336 * Interrupts the current utterance (whether played or rendered to file) and discards other 1337 * utterances in the queue. 1338 * 1339 * @return {@link #ERROR} or {@link #SUCCESS}. 1340 */ 1341 public int stop() { 1342 return runAction(new Action<Integer>() { 1343 @Override 1344 public Integer run(ITextToSpeechService service) throws RemoteException { 1345 return service.stop(getCallerIdentity()); 1346 } 1347 }, ERROR, "stop"); 1348 } 1349 1350 /** 1351 * Sets the speech rate. 1352 * 1353 * This has no effect on any pre-recorded speech. 1354 * 1355 * @param speechRate Speech rate. {@code 1.0} is the normal speech rate, 1356 * lower values slow down the speech ({@code 0.5} is half the normal speech rate), 1357 * greater values accelerate it ({@code 2.0} is twice the normal speech rate). 1358 * 1359 * @return {@link #ERROR} or {@link #SUCCESS}. 1360 */ 1361 public int setSpeechRate(float speechRate) { 1362 if (speechRate > 0.0f) { 1363 int intRate = (int)(speechRate * 100); 1364 if (intRate > 0) { 1365 synchronized (mStartLock) { 1366 mParams.putInt(Engine.KEY_PARAM_RATE, intRate); 1367 } 1368 return SUCCESS; 1369 } 1370 } 1371 return ERROR; 1372 } 1373 1374 /** 1375 * Sets the speech pitch for the TextToSpeech engine. 1376 * 1377 * This has no effect on any pre-recorded speech. 1378 * 1379 * @param pitch Speech pitch. {@code 1.0} is the normal pitch, 1380 * lower values lower the tone of the synthesized voice, 1381 * greater values increase it. 1382 * 1383 * @return {@link #ERROR} or {@link #SUCCESS}. 1384 */ 1385 public int setPitch(float pitch) { 1386 if (pitch > 0.0f) { 1387 int intPitch = (int)(pitch * 100); 1388 if (intPitch > 0) { 1389 synchronized (mStartLock) { 1390 mParams.putInt(Engine.KEY_PARAM_PITCH, intPitch); 1391 } 1392 return SUCCESS; 1393 } 1394 } 1395 return ERROR; 1396 } 1397 1398 /** 1399 * Sets the audio attributes to be used when speaking text or playing 1400 * back a file. 1401 * 1402 * @param audioAttributes Valid AudioAttributes instance. 1403 * 1404 * @return {@link #ERROR} or {@link #SUCCESS}. 1405 */ 1406 public int setAudioAttributes(AudioAttributes audioAttributes) { 1407 if (audioAttributes != null) { 1408 synchronized (mStartLock) { 1409 mParams.putParcelable(Engine.KEY_PARAM_AUDIO_ATTRIBUTES, 1410 audioAttributes); 1411 } 1412 return SUCCESS; 1413 } 1414 return ERROR; 1415 } 1416 1417 /** 1418 * @return the engine currently in use by this TextToSpeech instance. 1419 * @hide 1420 */ 1421 public String getCurrentEngine() { 1422 return mCurrentEngine; 1423 } 1424 1425 /** 1426 * Returns a Locale instance describing the language currently being used as the default 1427 * Text-to-speech language. 1428 * 1429 * The locale object returned by this method is NOT a valid one. It has identical form to the 1430 * one in {@link #getLanguage()}. Please refer to {@link #getLanguage()} for more information. 1431 * 1432 * @return language, country (if any) and variant (if any) used by the client stored in a 1433 * Locale instance, or {@code null} on error. 1434 * @deprecated As of API level 21, use <code>getDefaultVoice().getLocale()</code> ({@link 1435 * #getDefaultVoice()}) 1436 */ 1437 @Deprecated 1438 public Locale getDefaultLanguage() { 1439 return runAction(new Action<Locale>() { 1440 @Override 1441 public Locale run(ITextToSpeechService service) throws RemoteException { 1442 String[] defaultLanguage = service.getClientDefaultLanguage(); 1443 1444 return new Locale(defaultLanguage[0], defaultLanguage[1], defaultLanguage[2]); 1445 } 1446 }, null, "getDefaultLanguage"); 1447 } 1448 1449 /** 1450 * Sets the text-to-speech language. 1451 * The TTS engine will try to use the closest match to the specified 1452 * language as represented by the Locale, but there is no guarantee that the exact same Locale 1453 * will be used. Use {@link #isLanguageAvailable(Locale)} to check the level of support 1454 * before choosing the language to use for the next utterances. 1455 * 1456 * This method sets the current voice to the default one for the given Locale; 1457 * {@link #getVoice()} can be used to retrieve it. 1458 * 1459 * @param loc The locale describing the language to be used. 1460 * 1461 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1462 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1463 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1464 */ 1465 public int setLanguage(final Locale loc) { 1466 return runAction(new Action<Integer>() { 1467 @Override 1468 public Integer run(ITextToSpeechService service) throws RemoteException { 1469 if (loc == null) { 1470 return LANG_NOT_SUPPORTED; 1471 } 1472 String language = null, country = null; 1473 try { 1474 language = loc.getISO3Language(); 1475 } catch (MissingResourceException e) { 1476 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1477 return LANG_NOT_SUPPORTED; 1478 } 1479 1480 try { 1481 country = loc.getISO3Country(); 1482 } catch (MissingResourceException e) { 1483 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1484 return LANG_NOT_SUPPORTED; 1485 } 1486 1487 String variant = loc.getVariant(); 1488 1489 // As of API level 21, setLanguage is implemented using setVoice. 1490 // (which, in the default implementation, will call loadLanguage on the service 1491 // interface). 1492 1493 // Sanitize locale using isLanguageAvailable. 1494 int result = service.isLanguageAvailable(language, country, variant); 1495 if (result >= LANG_AVAILABLE) { 1496 // Get the default voice for the locale. 1497 String voiceName = service.getDefaultVoiceNameFor(language, country, variant); 1498 if (TextUtils.isEmpty(voiceName)) { 1499 Log.w(TAG, "Couldn't find the default voice for " + language + "-" + 1500 country + "-" + variant); 1501 return LANG_NOT_SUPPORTED; 1502 } 1503 1504 // Load it. 1505 if (service.loadVoice(getCallerIdentity(), voiceName) == TextToSpeech.ERROR) { 1506 Log.w(TAG, "The service claimed " + language + "-" + country + "-" 1507 + variant + " was available with voice name " + voiceName 1508 + " but loadVoice returned ERROR"); 1509 return LANG_NOT_SUPPORTED; 1510 } 1511 1512 // Set the language/country/variant of the voice, so #getLanguage will return 1513 // the currently set voice locale when called. 1514 Voice voice = getVoice(service, voiceName); 1515 if (voice == null) { 1516 Log.w(TAG, "getDefaultVoiceNameFor returned " + voiceName + " for locale " 1517 + language + "-" + country + "-" + variant 1518 + " but getVoice returns null"); 1519 return LANG_NOT_SUPPORTED; 1520 } 1521 String voiceLanguage = ""; 1522 try { 1523 voiceLanguage = voice.getLocale().getISO3Language(); 1524 } catch (MissingResourceException e) { 1525 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + 1526 voice.getLocale(), e); 1527 } 1528 1529 String voiceCountry = ""; 1530 try { 1531 voiceCountry = voice.getLocale().getISO3Country(); 1532 } catch (MissingResourceException e) { 1533 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + 1534 voice.getLocale(), e); 1535 } 1536 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, voiceName); 1537 mParams.putString(Engine.KEY_PARAM_LANGUAGE, voiceLanguage); 1538 mParams.putString(Engine.KEY_PARAM_COUNTRY, voiceCountry); 1539 mParams.putString(Engine.KEY_PARAM_VARIANT, voice.getLocale().getVariant()); 1540 } 1541 return result; 1542 } 1543 }, LANG_NOT_SUPPORTED, "setLanguage"); 1544 } 1545 1546 /** 1547 * Returns a Locale instance describing the language currently being used for synthesis 1548 * requests sent to the TextToSpeech engine. 1549 * 1550 * In Android 4.2 and before (API <= 17) this function returns the language that is currently 1551 * being used by the TTS engine. That is the last language set by this or any other 1552 * client by a {@link TextToSpeech#setLanguage} call to the same engine. 1553 * 1554 * In Android versions after 4.2 this function returns the language that is currently being 1555 * used for the synthesis requests sent from this client. That is the last language set 1556 * by a {@link TextToSpeech#setLanguage} call on this instance. 1557 * 1558 * If a voice is set (by {@link #setVoice(Voice)}), getLanguage will return the language of 1559 * the currently set voice. 1560 * 1561 * Please note that the Locale object returned by this method is NOT a valid Locale object. Its 1562 * language field contains a three-letter ISO 639-2/T code (where a proper Locale would use 1563 * a two-letter ISO 639-1 code), and the country field contains a three-letter ISO 3166 country 1564 * code (where a proper Locale would use a two-letter ISO 3166-1 code). 1565 * 1566 * @return language, country (if any) and variant (if any) used by the client stored in a 1567 * Locale instance, or {@code null} on error. 1568 * 1569 * @deprecated As of API level 21, please use <code>getVoice().getLocale()</code> 1570 * ({@link #getVoice()}). 1571 */ 1572 @Deprecated 1573 public Locale getLanguage() { 1574 return runAction(new Action<Locale>() { 1575 @Override 1576 public Locale run(ITextToSpeechService service) { 1577 /* No service call, but we're accessing mParams, hence need for 1578 wrapping it as an Action instance */ 1579 String lang = mParams.getString(Engine.KEY_PARAM_LANGUAGE, ""); 1580 String country = mParams.getString(Engine.KEY_PARAM_COUNTRY, ""); 1581 String variant = mParams.getString(Engine.KEY_PARAM_VARIANT, ""); 1582 return new Locale(lang, country, variant); 1583 } 1584 }, null, "getLanguage"); 1585 } 1586 1587 /** 1588 * Query the engine about the set of available languages. 1589 */ 1590 public Set<Locale> getAvailableLanguages() { 1591 return runAction(new Action<Set<Locale>>() { 1592 @Override 1593 public Set<Locale> run(ITextToSpeechService service) throws RemoteException { 1594 List<Voice> voices = service.getVoices(); 1595 if (voices == null) { 1596 return new HashSet<Locale>(); 1597 } 1598 HashSet<Locale> locales = new HashSet<Locale>(); 1599 for (Voice voice : voices) { 1600 locales.add(voice.getLocale()); 1601 } 1602 return locales; 1603 } 1604 }, null, "getAvailableLanguages"); 1605 } 1606 1607 /** 1608 * Query the engine about the set of available voices. 1609 * 1610 * Each TTS Engine can expose multiple voices for each locale, each with a different set of 1611 * features. 1612 * 1613 * @see #setVoice(Voice) 1614 * @see Voice 1615 */ 1616 public Set<Voice> getVoices() { 1617 return runAction(new Action<Set<Voice>>() { 1618 @Override 1619 public Set<Voice> run(ITextToSpeechService service) throws RemoteException { 1620 List<Voice> voices = service.getVoices(); 1621 return (voices != null) ? new HashSet<Voice>(voices) : new HashSet<Voice>(); 1622 } 1623 }, null, "getVoices"); 1624 } 1625 1626 /** 1627 * Sets the text-to-speech voice. 1628 * 1629 * @param voice One of objects returned by {@link #getVoices()}. 1630 * 1631 * @return {@link #ERROR} or {@link #SUCCESS}. 1632 * 1633 * @see #getVoices 1634 * @see Voice 1635 */ 1636 public int setVoice(final Voice voice) { 1637 return runAction(new Action<Integer>() { 1638 @Override 1639 public Integer run(ITextToSpeechService service) throws RemoteException { 1640 int result = service.loadVoice(getCallerIdentity(), voice.getName()); 1641 if (result == SUCCESS) { 1642 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, voice.getName()); 1643 1644 // Set the language/country/variant, so #getLanguage will return the voice 1645 // locale when called. 1646 String language = ""; 1647 try { 1648 language = voice.getLocale().getISO3Language(); 1649 } catch (MissingResourceException e) { 1650 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + 1651 voice.getLocale(), e); 1652 } 1653 1654 String country = ""; 1655 try { 1656 country = voice.getLocale().getISO3Country(); 1657 } catch (MissingResourceException e) { 1658 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + 1659 voice.getLocale(), e); 1660 } 1661 mParams.putString(Engine.KEY_PARAM_LANGUAGE, language); 1662 mParams.putString(Engine.KEY_PARAM_COUNTRY, country); 1663 mParams.putString(Engine.KEY_PARAM_VARIANT, voice.getLocale().getVariant()); 1664 } 1665 return result; 1666 } 1667 }, LANG_NOT_SUPPORTED, "setVoice"); 1668 } 1669 1670 /** 1671 * Returns a Voice instance describing the voice currently being used for synthesis 1672 * requests sent to the TextToSpeech engine. 1673 * 1674 * @return Voice instance used by the client, or {@code null} if not set or on error. 1675 * 1676 * @see #getVoices 1677 * @see #setVoice 1678 * @see Voice 1679 */ 1680 public Voice getVoice() { 1681 return runAction(new Action<Voice>() { 1682 @Override 1683 public Voice run(ITextToSpeechService service) throws RemoteException { 1684 String voiceName = mParams.getString(Engine.KEY_PARAM_VOICE_NAME, ""); 1685 if (TextUtils.isEmpty(voiceName)) { 1686 return null; 1687 } 1688 return getVoice(service, voiceName); 1689 } 1690 }, null, "getVoice"); 1691 } 1692 1693 1694 /** 1695 * Returns a Voice instance of the voice with the given voice name. 1696 * 1697 * @return Voice instance with the given voice name, or {@code null} if not set or on error. 1698 * 1699 * @see Voice 1700 */ 1701 private Voice getVoice(ITextToSpeechService service, String voiceName) throws RemoteException { 1702 List<Voice> voices = service.getVoices(); 1703 if (voices == null) { 1704 Log.w(TAG, "getVoices returned null"); 1705 return null; 1706 } 1707 for (Voice voice : voices) { 1708 if (voice.getName().equals(voiceName)) { 1709 return voice; 1710 } 1711 } 1712 Log.w(TAG, "Could not find voice " + voiceName + " in voice list"); 1713 return null; 1714 } 1715 1716 /** 1717 * Returns a Voice instance that's the default voice for the default Text-to-speech language. 1718 * @return The default voice instance for the default language, or {@code null} if not set or 1719 * on error. 1720 */ 1721 public Voice getDefaultVoice() { 1722 return runAction(new Action<Voice>() { 1723 @Override 1724 public Voice run(ITextToSpeechService service) throws RemoteException { 1725 1726 String[] defaultLanguage = service.getClientDefaultLanguage(); 1727 1728 if (defaultLanguage == null || defaultLanguage.length == 0) { 1729 Log.e(TAG, "service.getClientDefaultLanguage() returned empty array"); 1730 return null; 1731 } 1732 String language = defaultLanguage[0]; 1733 String country = (defaultLanguage.length > 1) ? defaultLanguage[1] : ""; 1734 String variant = (defaultLanguage.length > 2) ? defaultLanguage[2] : ""; 1735 1736 // Sanitize the locale using isLanguageAvailable. 1737 int result = service.isLanguageAvailable(language, country, variant); 1738 if (result < LANG_AVAILABLE) { 1739 // The default language is not supported. 1740 return null; 1741 } 1742 1743 // Get the default voice name 1744 String voiceName = service.getDefaultVoiceNameFor(language, country, variant); 1745 if (TextUtils.isEmpty(voiceName)) { 1746 return null; 1747 } 1748 1749 // Find it 1750 List<Voice> voices = service.getVoices(); 1751 if (voices == null) { 1752 return null; 1753 } 1754 for (Voice voice : voices) { 1755 if (voice.getName().equals(voiceName)) { 1756 return voice; 1757 } 1758 } 1759 return null; 1760 } 1761 }, null, "getDefaultVoice"); 1762 } 1763 1764 1765 1766 /** 1767 * Checks if the specified language as represented by the Locale is available and supported. 1768 * 1769 * @param loc The Locale describing the language to be used. 1770 * 1771 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1772 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1773 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1774 */ 1775 public int isLanguageAvailable(final Locale loc) { 1776 return runAction(new Action<Integer>() { 1777 @Override 1778 public Integer run(ITextToSpeechService service) throws RemoteException { 1779 String language = null, country = null; 1780 1781 try { 1782 language = loc.getISO3Language(); 1783 } catch (MissingResourceException e) { 1784 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1785 return LANG_NOT_SUPPORTED; 1786 } 1787 1788 try { 1789 country = loc.getISO3Country(); 1790 } catch (MissingResourceException e) { 1791 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1792 return LANG_NOT_SUPPORTED; 1793 } 1794 1795 return service.isLanguageAvailable(language, country, loc.getVariant()); 1796 } 1797 }, LANG_NOT_SUPPORTED, "isLanguageAvailable"); 1798 } 1799 1800 /** 1801 * Synthesizes the given text to a file using the specified parameters. 1802 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1803 * requests and then returns. The synthesis might not have finished (or even started!) at the 1804 * time when this method returns. In order to reliably detect errors during synthesis, 1805 * we recommend setting an utterance progress listener (see 1806 * {@link #setOnUtteranceProgressListener}). 1807 * 1808 * @param text The text that should be synthesized. No longer than 1809 * {@link #getMaxSpeechInputLength()} characters. 1810 * @param params Parameters for the request. Can be null. 1811 * Engine specific parameters may be passed in but the parameter keys 1812 * must be prefixed by the name of the engine they are intended for. For example 1813 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1814 * engine named "com.svox.pico" if it is being used. 1815 * @param file File to write the generated audio data to. 1816 * @param utteranceId An unique identifier for this request. 1817 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the synthesizeToFile operation. 1818 */ 1819 public int synthesizeToFile(final CharSequence text, final Bundle params, 1820 final File file, final String utteranceId) { 1821 return runAction(new Action<Integer>() { 1822 @Override 1823 public Integer run(ITextToSpeechService service) throws RemoteException { 1824 ParcelFileDescriptor fileDescriptor; 1825 int returnValue; 1826 try { 1827 if(file.exists() && !file.canWrite()) { 1828 Log.e(TAG, "Can't write to " + file); 1829 return ERROR; 1830 } 1831 fileDescriptor = ParcelFileDescriptor.open(file, 1832 ParcelFileDescriptor.MODE_WRITE_ONLY | 1833 ParcelFileDescriptor.MODE_CREATE | 1834 ParcelFileDescriptor.MODE_TRUNCATE); 1835 returnValue = service.synthesizeToFileDescriptor(getCallerIdentity(), text, 1836 fileDescriptor, getParams(params), utteranceId); 1837 fileDescriptor.close(); 1838 return returnValue; 1839 } catch (FileNotFoundException e) { 1840 Log.e(TAG, "Opening file " + file + " failed", e); 1841 return ERROR; 1842 } catch (IOException e) { 1843 Log.e(TAG, "Closing file " + file + " failed", e); 1844 return ERROR; 1845 } 1846 } 1847 }, ERROR, "synthesizeToFile"); 1848 } 1849 1850 /** 1851 * Synthesizes the given text to a file using the specified parameters. 1852 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1853 * requests and then returns. The synthesis might not have finished (or even started!) at the 1854 * time when this method returns. In order to reliably detect errors during synthesis, 1855 * we recommend setting an utterance progress listener (see 1856 * {@link #setOnUtteranceProgressListener}) and using the 1857 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1858 * 1859 * @param text The text that should be synthesized. No longer than 1860 * {@link #getMaxSpeechInputLength()} characters. 1861 * @param params Parameters for the request. Can be null. 1862 * Supported parameter names: 1863 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1864 * Engine specific parameters may be passed in but the parameter keys 1865 * must be prefixed by the name of the engine they are intended for. For example 1866 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1867 * engine named "com.svox.pico" if it is being used. 1868 * @param filename Absolute file filename to write the generated audio data to.It should be 1869 * something like "/sdcard/myappsounds/mysound.wav". 1870 * 1871 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the synthesizeToFile operation. 1872 * @deprecated As of API level 21, replaced by 1873 * {@link #synthesizeToFile(CharSequence, Bundle, File, String)}. 1874 */ 1875 @Deprecated 1876 public int synthesizeToFile(final String text, final HashMap<String, String> params, 1877 final String filename) { 1878 return synthesizeToFile(text, convertParamsHashMaptoBundle(params), 1879 new File(filename), params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1880 } 1881 1882 private Bundle convertParamsHashMaptoBundle(HashMap<String, String> params) { 1883 if (params != null && !params.isEmpty()) { 1884 Bundle bundle = new Bundle(); 1885 copyIntParam(bundle, params, Engine.KEY_PARAM_STREAM); 1886 copyIntParam(bundle, params, Engine.KEY_PARAM_SESSION_ID); 1887 copyStringParam(bundle, params, Engine.KEY_PARAM_UTTERANCE_ID); 1888 copyFloatParam(bundle, params, Engine.KEY_PARAM_VOLUME); 1889 copyFloatParam(bundle, params, Engine.KEY_PARAM_PAN); 1890 1891 // Copy feature strings defined by the framework. 1892 copyStringParam(bundle, params, Engine.KEY_FEATURE_NETWORK_SYNTHESIS); 1893 copyStringParam(bundle, params, Engine.KEY_FEATURE_EMBEDDED_SYNTHESIS); 1894 copyIntParam(bundle, params, Engine.KEY_FEATURE_NETWORK_TIMEOUT_MS); 1895 copyIntParam(bundle, params, Engine.KEY_FEATURE_NETWORK_RETRIES_COUNT); 1896 1897 // Copy over all parameters that start with the name of the 1898 // engine that we are currently connected to. The engine is 1899 // free to interpret them as it chooses. 1900 if (!TextUtils.isEmpty(mCurrentEngine)) { 1901 for (Map.Entry<String, String> entry : params.entrySet()) { 1902 final String key = entry.getKey(); 1903 if (key != null && key.startsWith(mCurrentEngine)) { 1904 bundle.putString(key, entry.getValue()); 1905 } 1906 } 1907 } 1908 1909 return bundle; 1910 } 1911 return null; 1912 } 1913 1914 private Bundle getParams(Bundle params) { 1915 if (params != null && !params.isEmpty()) { 1916 Bundle bundle = new Bundle(mParams); 1917 bundle.putAll(params); 1918 1919 verifyIntegerBundleParam(bundle, Engine.KEY_PARAM_STREAM); 1920 verifyIntegerBundleParam(bundle, Engine.KEY_PARAM_SESSION_ID); 1921 verifyStringBundleParam(bundle, Engine.KEY_PARAM_UTTERANCE_ID); 1922 verifyFloatBundleParam(bundle, Engine.KEY_PARAM_VOLUME); 1923 verifyFloatBundleParam(bundle, Engine.KEY_PARAM_PAN); 1924 1925 // Copy feature strings defined by the framework. 1926 verifyBooleanBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_SYNTHESIS); 1927 verifyBooleanBundleParam(bundle, Engine.KEY_FEATURE_EMBEDDED_SYNTHESIS); 1928 verifyIntegerBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_TIMEOUT_MS); 1929 verifyIntegerBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_RETRIES_COUNT); 1930 1931 return bundle; 1932 } else { 1933 return mParams; 1934 } 1935 } 1936 1937 private static boolean verifyIntegerBundleParam(Bundle bundle, String key) { 1938 if (bundle.containsKey(key)) { 1939 if (!(bundle.get(key) instanceof Integer || 1940 bundle.get(key) instanceof Long)) { 1941 bundle.remove(key); 1942 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1943 + " with invalid type. Should be an Integer or a Long"); 1944 return false; 1945 } 1946 } 1947 return true; 1948 } 1949 1950 private static boolean verifyStringBundleParam(Bundle bundle, String key) { 1951 if (bundle.containsKey(key)) { 1952 if (!(bundle.get(key) instanceof String)) { 1953 bundle.remove(key); 1954 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1955 + " with invalid type. Should be a String"); 1956 return false; 1957 } 1958 } 1959 return true; 1960 } 1961 1962 private static boolean verifyBooleanBundleParam(Bundle bundle, String key) { 1963 if (bundle.containsKey(key)) { 1964 if (!(bundle.get(key) instanceof Boolean || 1965 bundle.get(key) instanceof String)) { 1966 bundle.remove(key); 1967 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1968 + " with invalid type. Should be a Boolean or String"); 1969 return false; 1970 } 1971 } 1972 return true; 1973 } 1974 1975 1976 private static boolean verifyFloatBundleParam(Bundle bundle, String key) { 1977 if (bundle.containsKey(key)) { 1978 if (!(bundle.get(key) instanceof Float || 1979 bundle.get(key) instanceof Double)) { 1980 bundle.remove(key); 1981 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1982 + " with invalid type. Should be a Float or a Double"); 1983 return false; 1984 } 1985 } 1986 return true; 1987 } 1988 1989 private void copyStringParam(Bundle bundle, HashMap<String, String> params, String key) { 1990 String value = params.get(key); 1991 if (value != null) { 1992 bundle.putString(key, value); 1993 } 1994 } 1995 1996 private void copyIntParam(Bundle bundle, HashMap<String, String> params, String key) { 1997 String valueString = params.get(key); 1998 if (!TextUtils.isEmpty(valueString)) { 1999 try { 2000 int value = Integer.parseInt(valueString); 2001 bundle.putInt(key, value); 2002 } catch (NumberFormatException ex) { 2003 // don't set the value in the bundle 2004 } 2005 } 2006 } 2007 2008 private void copyFloatParam(Bundle bundle, HashMap<String, String> params, String key) { 2009 String valueString = params.get(key); 2010 if (!TextUtils.isEmpty(valueString)) { 2011 try { 2012 float value = Float.parseFloat(valueString); 2013 bundle.putFloat(key, value); 2014 } catch (NumberFormatException ex) { 2015 // don't set the value in the bundle 2016 } 2017 } 2018 } 2019 2020 /** 2021 * Sets the listener that will be notified when synthesis of an utterance completes. 2022 * 2023 * @param listener The listener to use. 2024 * 2025 * @return {@link #ERROR} or {@link #SUCCESS}. 2026 * 2027 * @deprecated Use {@link #setOnUtteranceProgressListener(UtteranceProgressListener)} 2028 * instead. 2029 */ 2030 @Deprecated 2031 public int setOnUtteranceCompletedListener(final OnUtteranceCompletedListener listener) { 2032 mUtteranceProgressListener = UtteranceProgressListener.from(listener); 2033 return TextToSpeech.SUCCESS; 2034 } 2035 2036 /** 2037 * Sets the listener that will be notified of various events related to the 2038 * synthesis of a given utterance. 2039 * 2040 * See {@link UtteranceProgressListener} and 2041 * {@link TextToSpeech.Engine#KEY_PARAM_UTTERANCE_ID}. 2042 * 2043 * @param listener the listener to use. 2044 * @return {@link #ERROR} or {@link #SUCCESS} 2045 */ 2046 public int setOnUtteranceProgressListener(UtteranceProgressListener listener) { 2047 mUtteranceProgressListener = listener; 2048 return TextToSpeech.SUCCESS; 2049 } 2050 2051 /** 2052 * Sets the TTS engine to use. 2053 * 2054 * @deprecated This doesn't inform callers when the TTS engine has been 2055 * initialized. {@link #TextToSpeech(Context, OnInitListener, String)} 2056 * can be used with the appropriate engine name. Also, there is no 2057 * guarantee that the engine specified will be loaded. If it isn't 2058 * installed or disabled, the user / system wide defaults will apply. 2059 * 2060 * @param enginePackageName The package name for the synthesis engine (e.g. "com.svox.pico") 2061 * 2062 * @return {@link #ERROR} or {@link #SUCCESS}. 2063 */ 2064 @Deprecated 2065 public int setEngineByPackageName(String enginePackageName) { 2066 mRequestedEngine = enginePackageName; 2067 return initTts(); 2068 } 2069 2070 /** 2071 * Gets the package name of the default speech synthesis engine. 2072 * 2073 * @return Package name of the TTS engine that the user has chosen 2074 * as their default. 2075 */ 2076 public String getDefaultEngine() { 2077 return mEnginesHelper.getDefaultEngine(); 2078 } 2079 2080 /** 2081 * Checks whether the user's settings should override settings requested 2082 * by the calling application. As of the Ice cream sandwich release, 2083 * user settings never forcibly override the app's settings. 2084 */ 2085 @Deprecated 2086 public boolean areDefaultsEnforced() { 2087 return false; 2088 } 2089 2090 /** 2091 * Gets a list of all installed TTS engines. 2092 * 2093 * @return A list of engine info objects. The list can be empty, but never {@code null}. 2094 */ 2095 public List<EngineInfo> getEngines() { 2096 return mEnginesHelper.getEngines(); 2097 } 2098 2099 private class Connection implements ServiceConnection { 2100 private ITextToSpeechService mService; 2101 2102 private SetupConnectionAsyncTask mOnSetupConnectionAsyncTask; 2103 2104 private boolean mEstablished; 2105 2106 private final ITextToSpeechCallback.Stub mCallback = 2107 new ITextToSpeechCallback.Stub() { 2108 public void onStop(String utteranceId, boolean isStarted) 2109 throws RemoteException { 2110 UtteranceProgressListener listener = mUtteranceProgressListener; 2111 if (listener != null) { 2112 listener.onStop(utteranceId, isStarted); 2113 } 2114 }; 2115 2116 @Override 2117 public void onSuccess(String utteranceId) { 2118 UtteranceProgressListener listener = mUtteranceProgressListener; 2119 if (listener != null) { 2120 listener.onDone(utteranceId); 2121 } 2122 } 2123 2124 @Override 2125 public void onError(String utteranceId, int errorCode) { 2126 UtteranceProgressListener listener = mUtteranceProgressListener; 2127 if (listener != null) { 2128 listener.onError(utteranceId); 2129 } 2130 } 2131 2132 @Override 2133 public void onStart(String utteranceId) { 2134 UtteranceProgressListener listener = mUtteranceProgressListener; 2135 if (listener != null) { 2136 listener.onStart(utteranceId); 2137 } 2138 } 2139 2140 @Override 2141 public void onBeginSynthesis( 2142 String utteranceId, 2143 int sampleRateInHz, 2144 int audioFormat, 2145 int channelCount) { 2146 UtteranceProgressListener listener = mUtteranceProgressListener; 2147 if (listener != null) { 2148 listener.onBeginSynthesis( 2149 utteranceId, sampleRateInHz, audioFormat, channelCount); 2150 } 2151 } 2152 2153 @Override 2154 public void onAudioAvailable(String utteranceId, byte[] audio) { 2155 UtteranceProgressListener listener = mUtteranceProgressListener; 2156 if (listener != null) { 2157 listener.onAudioAvailable(utteranceId, audio); 2158 } 2159 } 2160 2161 @Override 2162 public void onRangeStart(String utteranceId, int start, int end, int frame) { 2163 UtteranceProgressListener listener = mUtteranceProgressListener; 2164 if (listener != null) { 2165 listener.onRangeStart(utteranceId, start, end, frame); 2166 } 2167 } 2168 }; 2169 2170 private class SetupConnectionAsyncTask extends AsyncTask<Void, Void, Integer> { 2171 private final ComponentName mName; 2172 2173 public SetupConnectionAsyncTask(ComponentName name) { 2174 mName = name; 2175 } 2176 2177 @Override 2178 protected Integer doInBackground(Void... params) { 2179 synchronized(mStartLock) { 2180 if (isCancelled()) { 2181 return null; 2182 } 2183 2184 try { 2185 mService.setCallback(getCallerIdentity(), mCallback); 2186 2187 if (mParams.getString(Engine.KEY_PARAM_LANGUAGE) == null) { 2188 String[] defaultLanguage = mService.getClientDefaultLanguage(); 2189 mParams.putString(Engine.KEY_PARAM_LANGUAGE, defaultLanguage[0]); 2190 mParams.putString(Engine.KEY_PARAM_COUNTRY, defaultLanguage[1]); 2191 mParams.putString(Engine.KEY_PARAM_VARIANT, defaultLanguage[2]); 2192 2193 // Get the default voice for the locale. 2194 String defaultVoiceName = mService.getDefaultVoiceNameFor( 2195 defaultLanguage[0], defaultLanguage[1], defaultLanguage[2]); 2196 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, defaultVoiceName); 2197 } 2198 2199 Log.i(TAG, "Set up connection to " + mName); 2200 return SUCCESS; 2201 } catch (RemoteException re) { 2202 Log.e(TAG, "Error connecting to service, setCallback() failed"); 2203 return ERROR; 2204 } 2205 } 2206 } 2207 2208 @Override 2209 protected void onPostExecute(Integer result) { 2210 synchronized(mStartLock) { 2211 if (mOnSetupConnectionAsyncTask == this) { 2212 mOnSetupConnectionAsyncTask = null; 2213 } 2214 mEstablished = true; 2215 dispatchOnInit(result); 2216 } 2217 } 2218 } 2219 2220 @Override 2221 public void onServiceConnected(ComponentName name, IBinder service) { 2222 synchronized(mStartLock) { 2223 mConnectingServiceConnection = null; 2224 2225 Log.i(TAG, "Connected to " + name); 2226 2227 if (mOnSetupConnectionAsyncTask != null) { 2228 mOnSetupConnectionAsyncTask.cancel(false); 2229 } 2230 2231 mService = ITextToSpeechService.Stub.asInterface(service); 2232 mServiceConnection = Connection.this; 2233 2234 mEstablished = false; 2235 mOnSetupConnectionAsyncTask = new SetupConnectionAsyncTask(name); 2236 mOnSetupConnectionAsyncTask.execute(); 2237 } 2238 } 2239 2240 public IBinder getCallerIdentity() { 2241 return mCallback; 2242 } 2243 2244 /** 2245 * Clear connection related fields and cancel mOnServiceConnectedAsyncTask if set. 2246 * 2247 * @return true if we cancel mOnSetupConnectionAsyncTask in progress. 2248 */ 2249 private boolean clearServiceConnection() { 2250 synchronized(mStartLock) { 2251 boolean result = false; 2252 if (mOnSetupConnectionAsyncTask != null) { 2253 result = mOnSetupConnectionAsyncTask.cancel(false); 2254 mOnSetupConnectionAsyncTask = null; 2255 } 2256 2257 mService = null; 2258 // If this is the active connection, clear it 2259 if (mServiceConnection == this) { 2260 mServiceConnection = null; 2261 } 2262 return result; 2263 } 2264 } 2265 2266 @Override 2267 public void onServiceDisconnected(ComponentName name) { 2268 Log.i(TAG, "Asked to disconnect from " + name); 2269 if (clearServiceConnection()) { 2270 /* We need to protect against a rare case where engine 2271 * dies just after successful connection - and we process onServiceDisconnected 2272 * before OnServiceConnectedAsyncTask.onPostExecute. onServiceDisconnected cancels 2273 * OnServiceConnectedAsyncTask.onPostExecute and we don't call dispatchOnInit 2274 * with ERROR as argument. 2275 */ 2276 dispatchOnInit(ERROR); 2277 } 2278 } 2279 2280 public void disconnect() { 2281 mContext.unbindService(this); 2282 clearServiceConnection(); 2283 } 2284 2285 public boolean isEstablished() { 2286 return mService != null && mEstablished; 2287 } 2288 2289 public <R> R runAction(Action<R> action, R errorResult, String method, 2290 boolean reconnect, boolean onlyEstablishedConnection) { 2291 synchronized (mStartLock) { 2292 try { 2293 if (mService == null) { 2294 Log.w(TAG, method + " failed: not connected to TTS engine"); 2295 return errorResult; 2296 } 2297 if (onlyEstablishedConnection && !isEstablished()) { 2298 Log.w(TAG, method + " failed: TTS engine connection not fully set up"); 2299 return errorResult; 2300 } 2301 return action.run(mService); 2302 } catch (RemoteException ex) { 2303 Log.e(TAG, method + " failed", ex); 2304 if (reconnect) { 2305 disconnect(); 2306 initTts(); 2307 } 2308 return errorResult; 2309 } 2310 } 2311 } 2312 } 2313 2314 private interface Action<R> { 2315 R run(ITextToSpeechService service) throws RemoteException; 2316 } 2317 2318 /** 2319 * Information about an installed text-to-speech engine. 2320 * 2321 * @see TextToSpeech#getEngines 2322 */ 2323 public static class EngineInfo { 2324 /** 2325 * Engine package name.. 2326 */ 2327 public String name; 2328 /** 2329 * Localized label for the engine. 2330 */ 2331 public String label; 2332 /** 2333 * Icon for the engine. 2334 */ 2335 public int icon; 2336 /** 2337 * Whether this engine is a part of the system 2338 * image. 2339 * 2340 * @hide 2341 */ 2342 public boolean system; 2343 /** 2344 * The priority the engine declares for the the intent filter 2345 * {@code android.intent.action.TTS_SERVICE} 2346 * 2347 * @hide 2348 */ 2349 public int priority; 2350 2351 @Override 2352 public String toString() { 2353 return "EngineInfo{name=" + name + "}"; 2354 } 2355 2356 } 2357 2358 /** 2359 * Limit of length of input string passed to speak and synthesizeToFile. 2360 * 2361 * @see #speak 2362 * @see #synthesizeToFile 2363 */ 2364 public static int getMaxSpeechInputLength() { 2365 return 4000; 2366 } 2367 } 2368
