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(prefix = { "ERROR_" }, value = { 84 ERROR_SYNTHESIS, 85 ERROR_SERVICE, 86 ERROR_OUTPUT, 87 ERROR_NETWORK, 88 ERROR_NETWORK_TIMEOUT, 89 ERROR_INVALID_REQUEST, 90 ERROR_NOT_INSTALLED_YET 91 }) 92 @Retention(RetentionPolicy.SOURCE) 93 public @interface Error {} 94 95 /** 96 * Denotes a failure of a TTS engine to synthesize the given input. 97 */ 98 public static final int ERROR_SYNTHESIS = -3; 99 100 /** 101 * Denotes a failure of a TTS service. 102 */ 103 public static final int ERROR_SERVICE = -4; 104 105 /** 106 * Denotes a failure related to the output (audio device or a file). 107 */ 108 public static final int ERROR_OUTPUT = -5; 109 110 /** 111 * Denotes a failure caused by a network connectivity problems. 112 */ 113 public static final int ERROR_NETWORK = -6; 114 115 /** 116 * Denotes a failure caused by network timeout. 117 */ 118 public static final int ERROR_NETWORK_TIMEOUT = -7; 119 120 /** 121 * Denotes a failure caused by an invalid request. 122 */ 123 public static final int ERROR_INVALID_REQUEST = -8; 124 125 /** 126 * Denotes a failure caused by an unfinished download of the voice data. 127 * @see Engine#KEY_FEATURE_NOT_INSTALLED 128 */ 129 public static final int ERROR_NOT_INSTALLED_YET = -9; 130 131 /** 132 * Queue mode where all entries in the playback queue (media to be played 133 * and text to be synthesized) are dropped and replaced by the new entry. 134 * Queues are flushed with respect to a given calling app. Entries in the queue 135 * from other callees are not discarded. 136 */ 137 public static final int QUEUE_FLUSH = 0; 138 /** 139 * Queue mode where the new entry is added at the end of the playback queue. 140 */ 141 public static final int QUEUE_ADD = 1; 142 /** 143 * Queue mode where the entire playback queue is purged. This is different 144 * from {@link #QUEUE_FLUSH} in that all entries are purged, not just entries 145 * from a given caller. 146 * 147 * @hide 148 */ 149 static final int QUEUE_DESTROY = 2; 150 151 /** 152 * Denotes the language is available exactly as specified by the locale. 153 */ 154 public static final int LANG_COUNTRY_VAR_AVAILABLE = 2; 155 156 /** 157 * Denotes the language is available for the language and country specified 158 * by the locale, but not the variant. 159 */ 160 public static final int LANG_COUNTRY_AVAILABLE = 1; 161 162 /** 163 * Denotes the language is available for the language by the locale, 164 * but not the country and variant. 165 */ 166 public static final int LANG_AVAILABLE = 0; 167 168 /** 169 * Denotes the language data is missing. 170 */ 171 public static final int LANG_MISSING_DATA = -1; 172 173 /** 174 * Denotes the language is not supported. 175 */ 176 public static final int LANG_NOT_SUPPORTED = -2; 177 178 /** 179 * Broadcast Action: The TextToSpeech synthesizer has completed processing 180 * of all the text in the speech queue. 181 * 182 * Note that this notifies callers when the <b>engine</b> has finished has 183 * processing text data. Audio playback might not have completed (or even started) 184 * at this point. If you wish to be notified when this happens, see 185 * {@link OnUtteranceCompletedListener}. 186 */ 187 @SdkConstant(SdkConstantType.BROADCAST_INTENT_ACTION) 188 public static final String ACTION_TTS_QUEUE_PROCESSING_COMPLETED = 189 "android.speech.tts.TTS_QUEUE_PROCESSING_COMPLETED"; 190 191 /** 192 * Interface definition of a callback to be invoked indicating the completion of the 193 * TextToSpeech engine initialization. 194 */ 195 public interface OnInitListener { 196 /** 197 * Called to signal the completion of the TextToSpeech engine initialization. 198 * 199 * @param status {@link TextToSpeech#SUCCESS} or {@link TextToSpeech#ERROR}. 200 */ onInit(int status)201 void onInit(int status); 202 } 203 204 /** 205 * Listener that will be called when the TTS service has 206 * completed synthesizing an utterance. This is only called if the utterance 207 * has an utterance ID (see {@link TextToSpeech.Engine#KEY_PARAM_UTTERANCE_ID}). 208 * 209 * @deprecated Use {@link UtteranceProgressListener} instead. 210 */ 211 @Deprecated 212 public interface OnUtteranceCompletedListener { 213 /** 214 * Called when an utterance has been synthesized. 215 * 216 * @param utteranceId the identifier of the utterance. 217 */ onUtteranceCompleted(String utteranceId)218 void onUtteranceCompleted(String utteranceId); 219 } 220 221 /** 222 * Constants and parameter names for controlling text-to-speech. These include: 223 * 224 * <ul> 225 * <li> 226 * Intents to ask engine to install data or check its data and 227 * extras for a TTS engine's check data activity. 228 * </li> 229 * <li> 230 * Keys for the parameters passed with speak commands, e.g. 231 * {@link Engine#KEY_PARAM_UTTERANCE_ID}, {@link Engine#KEY_PARAM_STREAM}. 232 * </li> 233 * <li> 234 * A list of feature strings that engines might support, e.g 235 * {@link Engine#KEY_FEATURE_NETWORK_SYNTHESIS}. These values may be passed in to 236 * {@link TextToSpeech#speak} and {@link TextToSpeech#synthesizeToFile} to modify 237 * engine behaviour. The engine can be queried for the set of features it supports 238 * through {@link TextToSpeech#getFeatures(java.util.Locale)}. 239 * </li> 240 * </ul> 241 */ 242 public class Engine { 243 244 /** 245 * Default speech rate. 246 * @hide 247 */ 248 public static final int DEFAULT_RATE = 100; 249 250 /** 251 * Default pitch. 252 * @hide 253 */ 254 public static final int DEFAULT_PITCH = 100; 255 256 /** 257 * Default volume. 258 * @hide 259 */ 260 public static final float DEFAULT_VOLUME = 1.0f; 261 262 /** 263 * Default pan (centered). 264 * @hide 265 */ 266 public static final float DEFAULT_PAN = 0.0f; 267 268 /** 269 * Default value for {@link Settings.Secure#TTS_USE_DEFAULTS}. 270 * @hide 271 */ 272 public static final int USE_DEFAULTS = 0; // false 273 274 /** 275 * Package name of the default TTS engine. 276 * 277 * @hide 278 * @deprecated No longer in use, the default engine is determined by 279 * the sort order defined in {@link TtsEngines}. Note that 280 * this doesn't "break" anything because there is no guarantee that 281 * the engine specified below is installed on a given build, let 282 * alone be the default. 283 */ 284 @Deprecated 285 public static final String DEFAULT_ENGINE = "com.svox.pico"; 286 287 /** 288 * Default audio stream used when playing synthesized speech. 289 */ 290 public static final int DEFAULT_STREAM = AudioManager.STREAM_MUSIC; 291 292 /** 293 * Indicates success when checking the installation status of the resources used by the 294 * TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 295 */ 296 public static final int CHECK_VOICE_DATA_PASS = 1; 297 298 /** 299 * Indicates failure when checking the installation status of the resources used by the 300 * TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 301 */ 302 public static final int CHECK_VOICE_DATA_FAIL = 0; 303 304 /** 305 * Indicates erroneous data when checking the installation status of the resources used by 306 * the TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 307 * 308 * @deprecated Use CHECK_VOICE_DATA_FAIL instead. 309 */ 310 @Deprecated 311 public static final int CHECK_VOICE_DATA_BAD_DATA = -1; 312 313 /** 314 * Indicates missing resources when checking the installation status of the resources used 315 * by the TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 316 * 317 * @deprecated Use CHECK_VOICE_DATA_FAIL instead. 318 */ 319 @Deprecated 320 public static final int CHECK_VOICE_DATA_MISSING_DATA = -2; 321 322 /** 323 * Indicates missing storage volume when checking the installation status of the resources 324 * used by the TextToSpeech engine with the {@link #ACTION_CHECK_TTS_DATA} intent. 325 * 326 * @deprecated Use CHECK_VOICE_DATA_FAIL instead. 327 */ 328 @Deprecated 329 public static final int CHECK_VOICE_DATA_MISSING_VOLUME = -3; 330 331 /** 332 * Intent for starting a TTS service. Services that handle this intent must 333 * extend {@link TextToSpeechService}. Normal applications should not use this intent 334 * directly, instead they should talk to the TTS service using the the methods in this 335 * class. 336 */ 337 @SdkConstant(SdkConstantType.SERVICE_ACTION) 338 public static final String INTENT_ACTION_TTS_SERVICE = 339 "android.intent.action.TTS_SERVICE"; 340 341 /** 342 * Name under which a text to speech engine publishes information about itself. 343 * This meta-data should reference an XML resource containing a 344 * <code><{@link android.R.styleable#TextToSpeechEngine tts-engine}></code> 345 * tag. 346 */ 347 public static final String SERVICE_META_DATA = "android.speech.tts"; 348 349 // intents to ask engine to install data or check its data 350 /** 351 * Activity Action: Triggers the platform TextToSpeech engine to 352 * start the activity that installs the resource files on the device 353 * that are required for TTS to be operational. Since the installation 354 * of the data can be interrupted or declined by the user, the application 355 * shouldn't expect successful installation upon return from that intent, 356 * and if need be, should check installation status with 357 * {@link #ACTION_CHECK_TTS_DATA}. 358 */ 359 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 360 public static final String ACTION_INSTALL_TTS_DATA = 361 "android.speech.tts.engine.INSTALL_TTS_DATA"; 362 363 /** 364 * Broadcast Action: broadcast to signal the change in the list of available 365 * languages or/and their features. 366 */ 367 @SdkConstant(SdkConstantType.BROADCAST_INTENT_ACTION) 368 public static final String ACTION_TTS_DATA_INSTALLED = 369 "android.speech.tts.engine.TTS_DATA_INSTALLED"; 370 371 /** 372 * Activity Action: Starts the activity from the platform TextToSpeech 373 * engine to verify the proper installation and availability of the 374 * resource files on the system. Upon completion, the activity will 375 * return one of the following codes: 376 * {@link #CHECK_VOICE_DATA_PASS}, 377 * {@link #CHECK_VOICE_DATA_FAIL}, 378 * <p> Moreover, the data received in the activity result will contain the following 379 * fields: 380 * <ul> 381 * <li>{@link #EXTRA_AVAILABLE_VOICES} which contains an ArrayList<String> of all the 382 * available voices. The format of each voice is: lang-COUNTRY-variant where COUNTRY and 383 * variant are optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE").</li> 384 * <li>{@link #EXTRA_UNAVAILABLE_VOICES} which contains an ArrayList<String> of all the 385 * unavailable voices (ones that user can install). The format of each voice is: 386 * lang-COUNTRY-variant where COUNTRY and variant are optional (ie, "eng" or 387 * "eng-USA" or "eng-USA-FEMALE").</li> 388 * </ul> 389 */ 390 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 391 public static final String ACTION_CHECK_TTS_DATA = 392 "android.speech.tts.engine.CHECK_TTS_DATA"; 393 394 /** 395 * Activity intent for getting some sample text to use for demonstrating TTS. Specific 396 * locale have to be requested by passing following extra parameters: 397 * <ul> 398 * <li>language</li> 399 * <li>country</li> 400 * <li>variant</li> 401 * </ul> 402 * 403 * Upon completion, the activity result may contain the following fields: 404 * <ul> 405 * <li>{@link #EXTRA_SAMPLE_TEXT} which contains an String with sample text.</li> 406 * </ul> 407 */ 408 @SdkConstant(SdkConstantType.ACTIVITY_INTENT_ACTION) 409 public static final String ACTION_GET_SAMPLE_TEXT = 410 "android.speech.tts.engine.GET_SAMPLE_TEXT"; 411 412 /** 413 * Extra information received with the {@link #ACTION_GET_SAMPLE_TEXT} intent result where 414 * the TextToSpeech engine returns an String with sample text for requested voice 415 */ 416 public static final String EXTRA_SAMPLE_TEXT = "sampleText"; 417 418 419 // extras for a TTS engine's check data activity 420 /** 421 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 422 * the TextToSpeech engine returns an ArrayList<String> of all the available voices. 423 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 424 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 425 */ 426 public static final String EXTRA_AVAILABLE_VOICES = "availableVoices"; 427 428 /** 429 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 430 * the TextToSpeech engine returns an ArrayList<String> of all the unavailable voices. 431 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 432 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 433 */ 434 public static final String EXTRA_UNAVAILABLE_VOICES = "unavailableVoices"; 435 436 /** 437 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 438 * the TextToSpeech engine specifies the path to its resources. 439 * 440 * It may be used by language packages to find out where to put their data. 441 * 442 * @deprecated TTS engine implementation detail, this information has no use for 443 * text-to-speech API client. 444 */ 445 @Deprecated 446 public static final String EXTRA_VOICE_DATA_ROOT_DIRECTORY = "dataRoot"; 447 448 /** 449 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 450 * the TextToSpeech engine specifies the file names of its resources under the 451 * resource path. 452 * 453 * @deprecated TTS engine implementation detail, this information has no use for 454 * text-to-speech API client. 455 */ 456 @Deprecated 457 public static final String EXTRA_VOICE_DATA_FILES = "dataFiles"; 458 459 /** 460 * Extra information received with the {@link #ACTION_CHECK_TTS_DATA} intent result where 461 * the TextToSpeech engine specifies the locale associated with each resource file. 462 * 463 * @deprecated TTS engine implementation detail, this information has no use for 464 * text-to-speech API client. 465 */ 466 @Deprecated 467 public static final String EXTRA_VOICE_DATA_FILES_INFO = "dataFilesInfo"; 468 469 /** 470 * Extra information sent with the {@link #ACTION_CHECK_TTS_DATA} intent where the 471 * caller indicates to the TextToSpeech engine which specific sets of voice data to 472 * check for by sending an ArrayList<String> of the voices that are of interest. 473 * The format of each voice is: lang-COUNTRY-variant where COUNTRY and variant are 474 * optional (ie, "eng" or "eng-USA" or "eng-USA-FEMALE"). 475 * 476 * @deprecated Redundant functionality, checking for existence of specific sets of voice 477 * data can be done on client side. 478 */ 479 @Deprecated 480 public static final String EXTRA_CHECK_VOICE_DATA_FOR = "checkVoiceDataFor"; 481 482 // extras for a TTS engine's data installation 483 /** 484 * Extra information received with the {@link #ACTION_TTS_DATA_INSTALLED} intent result. 485 * It indicates whether the data files for the synthesis engine were successfully 486 * installed. The installation was initiated with the {@link #ACTION_INSTALL_TTS_DATA} 487 * intent. The possible values for this extra are 488 * {@link TextToSpeech#SUCCESS} and {@link TextToSpeech#ERROR}. 489 * 490 * @deprecated No longer in use. If client ise interested in information about what 491 * changed, is should send ACTION_CHECK_TTS_DATA intent to discover available voices. 492 */ 493 @Deprecated 494 public static final String EXTRA_TTS_DATA_INSTALLED = "dataInstalled"; 495 496 // keys for the parameters passed with speak commands. Hidden keys are used internally 497 // to maintain engine state for each TextToSpeech instance. 498 /** 499 * @hide 500 */ 501 public static final String KEY_PARAM_RATE = "rate"; 502 503 /** 504 * @hide 505 */ 506 public static final String KEY_PARAM_VOICE_NAME = "voiceName"; 507 508 /** 509 * @hide 510 */ 511 public static final String KEY_PARAM_LANGUAGE = "language"; 512 513 /** 514 * @hide 515 */ 516 public static final String KEY_PARAM_COUNTRY = "country"; 517 518 /** 519 * @hide 520 */ 521 public static final String KEY_PARAM_VARIANT = "variant"; 522 523 /** 524 * @hide 525 */ 526 public static final String KEY_PARAM_ENGINE = "engine"; 527 528 /** 529 * @hide 530 */ 531 public static final String KEY_PARAM_PITCH = "pitch"; 532 533 /** 534 * Parameter key to specify the audio stream type to be used when speaking text 535 * or playing back a file. The value should be one of the STREAM_ constants 536 * defined in {@link AudioManager}. 537 * 538 * @see TextToSpeech#speak(String, int, HashMap) 539 * @see TextToSpeech#playEarcon(String, int, HashMap) 540 */ 541 public static final String KEY_PARAM_STREAM = "streamType"; 542 543 /** 544 * Parameter key to specify the audio attributes to be used when 545 * speaking text or playing back a file. The value should be set 546 * using {@link TextToSpeech#setAudioAttributes(AudioAttributes)}. 547 * 548 * @see TextToSpeech#speak(String, int, HashMap) 549 * @see TextToSpeech#playEarcon(String, int, HashMap) 550 * @hide 551 */ 552 public static final String KEY_PARAM_AUDIO_ATTRIBUTES = "audioAttributes"; 553 554 /** 555 * Parameter key to identify an utterance in the 556 * {@link TextToSpeech.OnUtteranceCompletedListener} after text has been 557 * spoken, a file has been played back or a silence duration has elapsed. 558 * 559 * @see TextToSpeech#speak(String, int, HashMap) 560 * @see TextToSpeech#playEarcon(String, int, HashMap) 561 * @see TextToSpeech#synthesizeToFile(String, HashMap, String) 562 */ 563 public static final String KEY_PARAM_UTTERANCE_ID = "utteranceId"; 564 565 /** 566 * Parameter key to specify the speech volume relative to the current stream type 567 * volume used when speaking text. Volume is specified as a float ranging from 0 to 1 568 * where 0 is silence, and 1 is the maximum volume (the default behavior). 569 * 570 * @see TextToSpeech#speak(String, int, HashMap) 571 * @see TextToSpeech#playEarcon(String, int, HashMap) 572 */ 573 public static final String KEY_PARAM_VOLUME = "volume"; 574 575 /** 576 * Parameter key to specify how the speech is panned from left to right when speaking text. 577 * Pan is specified as a float ranging from -1 to +1 where -1 maps to a hard-left pan, 578 * 0 to center (the default behavior), and +1 to hard-right. 579 * 580 * @see TextToSpeech#speak(String, int, HashMap) 581 * @see TextToSpeech#playEarcon(String, int, HashMap) 582 */ 583 public static final String KEY_PARAM_PAN = "pan"; 584 585 /** 586 * Feature key for network synthesis. See {@link TextToSpeech#getFeatures(Locale)} 587 * for a description of how feature keys work. If set (and supported by the engine 588 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must 589 * use network based synthesis. 590 * 591 * @see TextToSpeech#speak(String, int, java.util.HashMap) 592 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 593 * @see TextToSpeech#getFeatures(java.util.Locale) 594 * 595 * @deprecated Starting from API level 21, to select network synthesis, call 596 * {@link TextToSpeech#getVoices()}, find a suitable network voice 597 * ({@link Voice#isNetworkConnectionRequired()}) and pass it 598 * to {@link TextToSpeech#setVoice(Voice)}. 599 */ 600 @Deprecated 601 public static final String KEY_FEATURE_NETWORK_SYNTHESIS = "networkTts"; 602 603 /** 604 * Feature key for embedded synthesis. See {@link TextToSpeech#getFeatures(Locale)} 605 * for a description of how feature keys work. If set and supported by the engine 606 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must synthesize 607 * text on-device (without making network requests). 608 * 609 * @see TextToSpeech#speak(String, int, java.util.HashMap) 610 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 611 * @see TextToSpeech#getFeatures(java.util.Locale) 612 613 * @deprecated Starting from API level 21, to select embedded synthesis, call 614 * ({@link TextToSpeech#getVoices()}, find a suitable embedded voice 615 * ({@link Voice#isNetworkConnectionRequired()}) and pass it 616 * to {@link TextToSpeech#setVoice(Voice)}). 617 */ 618 @Deprecated 619 public static final String KEY_FEATURE_EMBEDDED_SYNTHESIS = "embeddedTts"; 620 621 /** 622 * Parameter key to specify an audio session identifier (obtained from 623 * {@link AudioManager#generateAudioSessionId()}) that will be used by the request audio 624 * output. It can be used to associate one of the {@link android.media.audiofx.AudioEffect} 625 * objects with the synthesis (or earcon) output. 626 * 627 * @see TextToSpeech#speak(String, int, HashMap) 628 * @see TextToSpeech#playEarcon(String, int, HashMap) 629 */ 630 public static final String KEY_PARAM_SESSION_ID = "sessionId"; 631 632 /** 633 * Feature key that indicates that the voice may need to download additional data to be fully 634 * functional. The download will be triggered by calling 635 * {@link TextToSpeech#setVoice(Voice)} or {@link TextToSpeech#setLanguage(Locale)}. 636 * Until download is complete, each synthesis request will either report 637 * {@link TextToSpeech#ERROR_NOT_INSTALLED_YET} error, or use a different voice to synthesize 638 * the request. This feature should NOT be used as a key of a request parameter. 639 * 640 * @see TextToSpeech#getFeatures(java.util.Locale) 641 * @see Voice#getFeatures() 642 */ 643 public static final String KEY_FEATURE_NOT_INSTALLED = "notInstalled"; 644 645 /** 646 * Feature key that indicate that a network timeout can be set for the request. If set and 647 * supported as per {@link TextToSpeech#getFeatures(Locale)} or {@link Voice#getFeatures()}, 648 * it can be used as request parameter to set the maximum allowed time for a single 649 * request attempt, in milliseconds, before synthesis fails. When used as a key of 650 * a request parameter, its value should be a string with an integer value. 651 * 652 * @see TextToSpeech#getFeatures(java.util.Locale) 653 * @see Voice#getFeatures() 654 */ 655 public static final String KEY_FEATURE_NETWORK_TIMEOUT_MS = "networkTimeoutMs"; 656 657 /** 658 * Feature key that indicates that network request retries count can be set for the request. 659 * If set and supported as per {@link TextToSpeech#getFeatures(Locale)} or 660 * {@link Voice#getFeatures()}, it can be used as a request parameter to set the 661 * number of network request retries that are attempted in case of failure. When used as 662 * a key of a request parameter, its value should be a string with an integer value. 663 * 664 * @see TextToSpeech#getFeatures(java.util.Locale) 665 * @see Voice#getFeatures() 666 */ 667 public static final String KEY_FEATURE_NETWORK_RETRIES_COUNT = "networkRetriesCount"; 668 } 669 670 private final Context mContext; 671 private Connection mConnectingServiceConnection; 672 private Connection mServiceConnection; 673 private OnInitListener mInitListener; 674 // Written from an unspecified application thread, read from 675 // a binder thread. 676 @Nullable private volatile UtteranceProgressListener mUtteranceProgressListener; 677 private final Object mStartLock = new Object(); 678 679 private String mRequestedEngine; 680 // Whether to initialize this TTS object with the default engine, 681 // if the requested engine is not available. Valid only if mRequestedEngine 682 // is not null. Used only for testing, though potentially useful API wise 683 // too. 684 private final boolean mUseFallback; 685 private final Map<String, Uri> mEarcons; 686 private final Map<CharSequence, Uri> mUtterances; 687 private final Bundle mParams = new Bundle(); 688 private final TtsEngines mEnginesHelper; 689 private volatile String mCurrentEngine = null; 690 691 /** 692 * The constructor for the TextToSpeech class, using the default TTS engine. 693 * This will also initialize the associated TextToSpeech engine if it isn't already running. 694 * 695 * @param context 696 * The context this instance is running in. 697 * @param listener 698 * The {@link TextToSpeech.OnInitListener} that will be called when the 699 * TextToSpeech engine has initialized. In a case of a failure the listener 700 * may be called immediately, before TextToSpeech instance is fully constructed. 701 */ TextToSpeech(Context context, OnInitListener listener)702 public TextToSpeech(Context context, OnInitListener listener) { 703 this(context, listener, null); 704 } 705 706 /** 707 * The constructor for the TextToSpeech class, using the given TTS engine. 708 * This will also initialize the associated TextToSpeech engine if it isn't already running. 709 * 710 * @param context 711 * The context this instance is running in. 712 * @param listener 713 * The {@link TextToSpeech.OnInitListener} that will be called when the 714 * TextToSpeech engine has initialized. In a case of a failure the listener 715 * may be called immediately, before TextToSpeech instance is fully constructed. 716 * @param engine Package name of the TTS engine to use. 717 */ TextToSpeech(Context context, OnInitListener listener, String engine)718 public TextToSpeech(Context context, OnInitListener listener, String engine) { 719 this(context, listener, engine, null, true); 720 } 721 722 /** 723 * Used by the framework to instantiate TextToSpeech objects with a supplied 724 * package name, instead of using {@link android.content.Context#getPackageName()} 725 * 726 * @hide 727 */ TextToSpeech(Context context, OnInitListener listener, String engine, String packageName, boolean useFallback)728 public TextToSpeech(Context context, OnInitListener listener, String engine, 729 String packageName, boolean useFallback) { 730 mContext = context; 731 mInitListener = listener; 732 mRequestedEngine = engine; 733 mUseFallback = useFallback; 734 735 mEarcons = new HashMap<String, Uri>(); 736 mUtterances = new HashMap<CharSequence, Uri>(); 737 mUtteranceProgressListener = null; 738 739 mEnginesHelper = new TtsEngines(mContext); 740 initTts(); 741 } 742 runActionNoReconnect(Action<R> action, R errorResult, String method, boolean onlyEstablishedConnection)743 private <R> R runActionNoReconnect(Action<R> action, R errorResult, String method, 744 boolean onlyEstablishedConnection) { 745 return runAction(action, errorResult, method, false, onlyEstablishedConnection); 746 } 747 runAction(Action<R> action, R errorResult, String method)748 private <R> R runAction(Action<R> action, R errorResult, String method) { 749 return runAction(action, errorResult, method, true, true); 750 } 751 runAction(Action<R> action, R errorResult, String method, boolean reconnect, boolean onlyEstablishedConnection)752 private <R> R runAction(Action<R> action, R errorResult, String method, 753 boolean reconnect, boolean onlyEstablishedConnection) { 754 synchronized (mStartLock) { 755 if (mServiceConnection == null) { 756 Log.w(TAG, method + " failed: not bound to TTS engine"); 757 return errorResult; 758 } 759 return mServiceConnection.runAction(action, errorResult, method, reconnect, 760 onlyEstablishedConnection); 761 } 762 } 763 initTts()764 private int initTts() { 765 // Step 1: Try connecting to the engine that was requested. 766 if (mRequestedEngine != null) { 767 if (mEnginesHelper.isEngineInstalled(mRequestedEngine)) { 768 if (connectToEngine(mRequestedEngine)) { 769 mCurrentEngine = mRequestedEngine; 770 return SUCCESS; 771 } else if (!mUseFallback) { 772 mCurrentEngine = null; 773 dispatchOnInit(ERROR); 774 return ERROR; 775 } 776 } else if (!mUseFallback) { 777 Log.i(TAG, "Requested engine not installed: " + mRequestedEngine); 778 mCurrentEngine = null; 779 dispatchOnInit(ERROR); 780 return ERROR; 781 } 782 } 783 784 // Step 2: Try connecting to the user's default engine. 785 final String defaultEngine = getDefaultEngine(); 786 if (defaultEngine != null && !defaultEngine.equals(mRequestedEngine)) { 787 if (connectToEngine(defaultEngine)) { 788 mCurrentEngine = defaultEngine; 789 return SUCCESS; 790 } 791 } 792 793 // Step 3: Try connecting to the highest ranked engine in the 794 // system. 795 final String highestRanked = mEnginesHelper.getHighestRankedEngineName(); 796 if (highestRanked != null && !highestRanked.equals(mRequestedEngine) && 797 !highestRanked.equals(defaultEngine)) { 798 if (connectToEngine(highestRanked)) { 799 mCurrentEngine = highestRanked; 800 return SUCCESS; 801 } 802 } 803 804 // NOTE: The API currently does not allow the caller to query whether 805 // they are actually connected to any engine. This might fail for various 806 // reasons like if the user disables all her TTS engines. 807 808 mCurrentEngine = null; 809 dispatchOnInit(ERROR); 810 return ERROR; 811 } 812 connectToEngine(String engine)813 private boolean connectToEngine(String engine) { 814 Connection connection = new Connection(); 815 Intent intent = new Intent(Engine.INTENT_ACTION_TTS_SERVICE); 816 intent.setPackage(engine); 817 boolean bound = mContext.bindService(intent, connection, Context.BIND_AUTO_CREATE); 818 if (!bound) { 819 Log.e(TAG, "Failed to bind to " + engine); 820 return false; 821 } else { 822 Log.i(TAG, "Sucessfully bound to " + engine); 823 mConnectingServiceConnection = connection; 824 return true; 825 } 826 } 827 dispatchOnInit(int result)828 private void dispatchOnInit(int result) { 829 synchronized (mStartLock) { 830 if (mInitListener != null) { 831 mInitListener.onInit(result); 832 mInitListener = null; 833 } 834 } 835 } 836 getCallerIdentity()837 private IBinder getCallerIdentity() { 838 return mServiceConnection.getCallerIdentity(); 839 } 840 841 /** 842 * Releases the resources used by the TextToSpeech engine. 843 * It is good practice for instance to call this method in the onDestroy() method of an Activity 844 * so the TextToSpeech engine can be cleanly stopped. 845 */ shutdown()846 public void shutdown() { 847 // Special case, we are asked to shutdown connection that did finalize its connection. 848 synchronized (mStartLock) { 849 if (mConnectingServiceConnection != null) { 850 mContext.unbindService(mConnectingServiceConnection); 851 mConnectingServiceConnection = null; 852 return; 853 } 854 } 855 856 // Post connection case 857 runActionNoReconnect(new Action<Void>() { 858 @Override 859 public Void run(ITextToSpeechService service) throws RemoteException { 860 service.setCallback(getCallerIdentity(), null); 861 service.stop(getCallerIdentity()); 862 mServiceConnection.disconnect(); 863 // Context#unbindService does not result in a call to 864 // ServiceConnection#onServiceDisconnected. As a result, the 865 // service ends up being destroyed (if there are no other open 866 // connections to it) but the process lives on and the 867 // ServiceConnection continues to refer to the destroyed service. 868 // 869 // This leads to tons of log spam about SynthThread being dead. 870 mServiceConnection = null; 871 mCurrentEngine = null; 872 return null; 873 } 874 }, null, "shutdown", false); 875 } 876 877 /** 878 * Adds a mapping between a string of text and a sound resource in a 879 * package. After a call to this method, subsequent calls to 880 * {@link #speak(String, int, HashMap)} will play the specified sound resource 881 * if it is available, or synthesize the text it is missing. 882 * 883 * @param text 884 * The string of text. Example: <code>"south_south_east"</code> 885 * 886 * @param packagename 887 * Pass the packagename of the application that contains the 888 * resource. If the resource is in your own application (this is 889 * the most common case), then put the packagename of your 890 * application here.<br/> 891 * Example: <b>"com.google.marvin.compass"</b><br/> 892 * The packagename can be found in the AndroidManifest.xml of 893 * your application. 894 * <p> 895 * <code><manifest xmlns:android="..." 896 * package="<b>com.google.marvin.compass</b>"></code> 897 * </p> 898 * 899 * @param resourceId 900 * Example: <code>R.raw.south_south_east</code> 901 * 902 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 903 */ addSpeech(String text, String packagename, @RawRes int resourceId)904 public int addSpeech(String text, String packagename, @RawRes int resourceId) { 905 synchronized (mStartLock) { 906 mUtterances.put(text, makeResourceUri(packagename, resourceId)); 907 return SUCCESS; 908 } 909 } 910 911 /** 912 * Adds a mapping between a CharSequence (may be spanned with TtsSpans) of text 913 * and a sound resource in a package. After a call to this method, subsequent calls to 914 * {@link #speak(String, int, HashMap)} will play the specified sound resource 915 * if it is available, or synthesize the text it is missing. 916 * 917 * @param text 918 * The string of text. Example: <code>"south_south_east"</code> 919 * 920 * @param packagename 921 * Pass the packagename of the application that contains the 922 * resource. If the resource is in your own application (this is 923 * the most common case), then put the packagename of your 924 * application here.<br/> 925 * Example: <b>"com.google.marvin.compass"</b><br/> 926 * The packagename can be found in the AndroidManifest.xml of 927 * your application. 928 * <p> 929 * <code><manifest xmlns:android="..." 930 * package="<b>com.google.marvin.compass</b>"></code> 931 * </p> 932 * 933 * @param resourceId 934 * Example: <code>R.raw.south_south_east</code> 935 * 936 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 937 */ addSpeech(CharSequence text, String packagename, @RawRes int resourceId)938 public int addSpeech(CharSequence text, String packagename, @RawRes int resourceId) { 939 synchronized (mStartLock) { 940 mUtterances.put(text, makeResourceUri(packagename, resourceId)); 941 return SUCCESS; 942 } 943 } 944 945 /** 946 * Adds a mapping between a string of text and a sound file. Using this, it 947 * is possible to add custom pronounciations for a string of text. 948 * After a call to this method, subsequent calls to {@link #speak(String, int, HashMap)} 949 * will play the specified sound resource if it is available, or synthesize the text it is 950 * missing. 951 * 952 * @param text 953 * The string of text. Example: <code>"south_south_east"</code> 954 * @param filename 955 * The full path to the sound file (for example: 956 * "/sdcard/mysounds/hello.wav") 957 * 958 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 959 */ addSpeech(String text, String filename)960 public int addSpeech(String text, String filename) { 961 synchronized (mStartLock) { 962 mUtterances.put(text, Uri.parse(filename)); 963 return SUCCESS; 964 } 965 } 966 967 /** 968 * Adds a mapping between a CharSequence (may be spanned with TtsSpans and a sound file. 969 * Using this, it is possible to add custom pronounciations for a string of text. 970 * After a call to this method, subsequent calls to {@link #speak(String, int, HashMap)} 971 * will play the specified sound resource if it is available, or synthesize the text it is 972 * missing. 973 * 974 * @param text 975 * The string of text. Example: <code>"south_south_east"</code> 976 * @param file 977 * File object pointing to the sound file. 978 * 979 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 980 */ addSpeech(CharSequence text, File file)981 public int addSpeech(CharSequence text, File file) { 982 synchronized (mStartLock) { 983 mUtterances.put(text, Uri.fromFile(file)); 984 return SUCCESS; 985 } 986 } 987 988 /** 989 * Adds a mapping between a string of text and a sound resource in a 990 * package. Use this to add custom earcons. 991 * 992 * @see #playEarcon(String, int, HashMap) 993 * 994 * @param earcon The name of the earcon. 995 * Example: <code>"[tick]"</code><br/> 996 * 997 * @param packagename 998 * the package name of the application that contains the 999 * resource. This can for instance be the package name of your own application. 1000 * Example: <b>"com.google.marvin.compass"</b><br/> 1001 * The package name can be found in the AndroidManifest.xml of 1002 * the application containing the resource. 1003 * <p> 1004 * <code><manifest xmlns:android="..." 1005 * package="<b>com.google.marvin.compass</b>"></code> 1006 * </p> 1007 * 1008 * @param resourceId 1009 * Example: <code>R.raw.tick_snd</code> 1010 * 1011 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 1012 */ addEarcon(String earcon, String packagename, @RawRes int resourceId)1013 public int addEarcon(String earcon, String packagename, @RawRes int resourceId) { 1014 synchronized(mStartLock) { 1015 mEarcons.put(earcon, makeResourceUri(packagename, resourceId)); 1016 return SUCCESS; 1017 } 1018 } 1019 1020 /** 1021 * Adds a mapping between a string of text and a sound file. 1022 * Use this to add custom earcons. 1023 * 1024 * @see #playEarcon(String, int, HashMap) 1025 * 1026 * @param earcon 1027 * The name of the earcon. 1028 * Example: <code>"[tick]"</code> 1029 * @param filename 1030 * The full path to the sound file (for example: 1031 * "/sdcard/mysounds/tick.wav") 1032 * 1033 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 1034 * 1035 * @deprecated As of API level 21, replaced by 1036 * {@link #addEarcon(String, File)}. 1037 */ 1038 @Deprecated addEarcon(String earcon, String filename)1039 public int addEarcon(String earcon, String filename) { 1040 synchronized(mStartLock) { 1041 mEarcons.put(earcon, Uri.parse(filename)); 1042 return SUCCESS; 1043 } 1044 } 1045 1046 /** 1047 * Adds a mapping between a string of text and a sound file. 1048 * Use this to add custom earcons. 1049 * 1050 * @see #playEarcon(String, int, HashMap) 1051 * 1052 * @param earcon 1053 * The name of the earcon. 1054 * Example: <code>"[tick]"</code> 1055 * @param file 1056 * File object pointing to the sound file. 1057 * 1058 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 1059 */ addEarcon(String earcon, File file)1060 public int addEarcon(String earcon, File file) { 1061 synchronized(mStartLock) { 1062 mEarcons.put(earcon, Uri.fromFile(file)); 1063 return SUCCESS; 1064 } 1065 } 1066 makeResourceUri(String packageName, int resourceId)1067 private Uri makeResourceUri(String packageName, int resourceId) { 1068 return new Uri.Builder() 1069 .scheme(ContentResolver.SCHEME_ANDROID_RESOURCE) 1070 .encodedAuthority(packageName) 1071 .appendEncodedPath(String.valueOf(resourceId)) 1072 .build(); 1073 } 1074 1075 /** 1076 * Speaks the text using the specified queuing strategy and speech parameters, the text may 1077 * be spanned with TtsSpans. 1078 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1079 * requests and then returns. The synthesis might not have finished (or even started!) at the 1080 * time when this method returns. In order to reliably detect errors during synthesis, 1081 * we recommend setting an utterance progress listener (see 1082 * {@link #setOnUtteranceProgressListener}) and using the 1083 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1084 * 1085 * @param text The string of text to be spoken. No longer than 1086 * {@link #getMaxSpeechInputLength()} characters. 1087 * @param queueMode The queuing strategy to use, {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1088 * @param params Parameters for the request. Can be null. 1089 * Supported parameter names: 1090 * {@link Engine#KEY_PARAM_STREAM}, 1091 * {@link Engine#KEY_PARAM_VOLUME}, 1092 * {@link Engine#KEY_PARAM_PAN}. 1093 * Engine specific parameters may be passed in but the parameter keys 1094 * must be prefixed by the name of the engine they are intended for. For example 1095 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1096 * engine named "com.svox.pico" if it is being used. 1097 * @param utteranceId An unique identifier for this request. 1098 * 1099 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the speak operation. 1100 */ speak(final CharSequence text, final int queueMode, final Bundle params, final String utteranceId)1101 public int speak(final CharSequence text, 1102 final int queueMode, 1103 final Bundle params, 1104 final String utteranceId) { 1105 return runAction(new Action<Integer>() { 1106 @Override 1107 public Integer run(ITextToSpeechService service) throws RemoteException { 1108 Uri utteranceUri = mUtterances.get(text); 1109 if (utteranceUri != null) { 1110 return service.playAudio(getCallerIdentity(), utteranceUri, queueMode, 1111 getParams(params), utteranceId); 1112 } else { 1113 return service.speak(getCallerIdentity(), text, queueMode, getParams(params), 1114 utteranceId); 1115 } 1116 } 1117 }, ERROR, "speak"); 1118 } 1119 1120 /** 1121 * Speaks the string using the specified queuing strategy and speech parameters. 1122 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1123 * requests and then returns. The synthesis might not have finished (or even started!) at the 1124 * time when this method returns. In order to reliably detect errors during synthesis, 1125 * we recommend setting an utterance progress listener (see 1126 * {@link #setOnUtteranceProgressListener}) and using the 1127 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1128 * 1129 * @param text The string of text to be spoken. No longer than 1130 * {@link #getMaxSpeechInputLength()} characters. 1131 * @param queueMode The queuing strategy to use, {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1132 * @param params Parameters for the request. Can be null. 1133 * Supported parameter names: 1134 * {@link Engine#KEY_PARAM_STREAM}, 1135 * {@link Engine#KEY_PARAM_UTTERANCE_ID}, 1136 * {@link Engine#KEY_PARAM_VOLUME}, 1137 * {@link Engine#KEY_PARAM_PAN}. 1138 * Engine specific parameters may be passed in but the parameter keys 1139 * must be prefixed by the name of the engine they are intended for. For example 1140 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1141 * engine named "com.svox.pico" if it is being used. 1142 * 1143 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the speak operation. 1144 * @deprecated As of API level 21, replaced by 1145 * {@link #speak(CharSequence, int, Bundle, String)}. 1146 */ 1147 @Deprecated 1148 public int speak(final String text, final int queueMode, final HashMap<String, String> params) { 1149 return speak(text, queueMode, convertParamsHashMaptoBundle(params), 1150 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1151 } 1152 1153 /** 1154 * Plays the earcon using the specified queueing mode and parameters. 1155 * The earcon must already have been added with {@link #addEarcon(String, String)} or 1156 * {@link #addEarcon(String, String, int)}. 1157 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1158 * requests and then returns. The synthesis might not have finished (or even started!) at the 1159 * time when this method returns. In order to reliably detect errors during synthesis, 1160 * we recommend setting an utterance progress listener (see 1161 * {@link #setOnUtteranceProgressListener}) and using the 1162 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1163 * 1164 * @param earcon The earcon that should be played 1165 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1166 * @param params Parameters for the request. Can be null. 1167 * Supported parameter names: 1168 * {@link Engine#KEY_PARAM_STREAM}, 1169 * Engine specific parameters may be passed in but the parameter keys 1170 * must be prefixed by the name of the engine they are intended for. For example 1171 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1172 * engine named "com.svox.pico" if it is being used. 1173 * 1174 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playEarcon operation. 1175 */ 1176 public int playEarcon(final String earcon, final int queueMode, 1177 final Bundle params, final String utteranceId) { 1178 return runAction(new Action<Integer>() { 1179 @Override 1180 public Integer run(ITextToSpeechService service) throws RemoteException { 1181 Uri earconUri = mEarcons.get(earcon); 1182 if (earconUri == null) { 1183 return ERROR; 1184 } 1185 return service.playAudio(getCallerIdentity(), earconUri, queueMode, 1186 getParams(params), utteranceId); 1187 } 1188 }, ERROR, "playEarcon"); 1189 } 1190 1191 /** 1192 * Plays the earcon using the specified queueing mode and parameters. 1193 * The earcon must already have been added with {@link #addEarcon(String, String)} or 1194 * {@link #addEarcon(String, String, int)}. 1195 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1196 * requests and then returns. The synthesis might not have finished (or even started!) at the 1197 * time when this method returns. In order to reliably detect errors during synthesis, 1198 * we recommend setting an utterance progress listener (see 1199 * {@link #setOnUtteranceProgressListener}) and using the 1200 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1201 * 1202 * @param earcon The earcon that should be played 1203 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1204 * @param params Parameters for the request. Can be null. 1205 * Supported parameter names: 1206 * {@link Engine#KEY_PARAM_STREAM}, 1207 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1208 * Engine specific parameters may be passed in but the parameter keys 1209 * must be prefixed by the name of the engine they are intended for. For example 1210 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1211 * engine named "com.svox.pico" if it is being used. 1212 * 1213 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playEarcon operation. 1214 * @deprecated As of API level 21, replaced by 1215 * {@link #playEarcon(String, int, Bundle, String)}. 1216 */ 1217 @Deprecated 1218 public int playEarcon(final String earcon, final int queueMode, 1219 final HashMap<String, String> params) { 1220 return playEarcon(earcon, queueMode, convertParamsHashMaptoBundle(params), 1221 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1222 } 1223 1224 /** 1225 * Plays silence for the specified amount of time using the specified 1226 * queue mode. 1227 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1228 * requests and then returns. The synthesis might not have finished (or even started!) at the 1229 * time when this method returns. In order to reliably detect errors during synthesis, 1230 * we recommend setting an utterance progress listener (see 1231 * {@link #setOnUtteranceProgressListener}) and using the 1232 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1233 * 1234 * @param durationInMs The duration of the silence. 1235 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1236 * @param utteranceId An unique identifier for this request. 1237 * 1238 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playSilentUtterance operation. 1239 */ 1240 public int playSilentUtterance(final long durationInMs, final int queueMode, 1241 final String utteranceId) { 1242 return runAction(new Action<Integer>() { 1243 @Override 1244 public Integer run(ITextToSpeechService service) throws RemoteException { 1245 return service.playSilence(getCallerIdentity(), durationInMs, 1246 queueMode, utteranceId); 1247 } 1248 }, ERROR, "playSilentUtterance"); 1249 } 1250 1251 /** 1252 * Plays silence for the specified amount of time using the specified 1253 * queue mode. 1254 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1255 * requests and then returns. The synthesis might not have finished (or even started!) at the 1256 * time when this method returns. In order to reliably detect errors during synthesis, 1257 * we recommend setting an utterance progress listener (see 1258 * {@link #setOnUtteranceProgressListener}) and using the 1259 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1260 * 1261 * @param durationInMs The duration of the silence. 1262 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1263 * @param params Parameters for the request. Can be null. 1264 * Supported parameter names: 1265 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1266 * Engine specific parameters may be passed in but the parameter keys 1267 * must be prefixed by the name of the engine they are intended for. For example 1268 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1269 * engine named "com.svox.pico" if it is being used. 1270 * 1271 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playSilence operation. 1272 * @deprecated As of API level 21, replaced by 1273 * {@link #playSilentUtterance(long, int, String)}. 1274 */ 1275 @Deprecated 1276 public int playSilence(final long durationInMs, final int queueMode, 1277 final HashMap<String, String> params) { 1278 return playSilentUtterance(durationInMs, queueMode, 1279 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1280 } 1281 1282 /** 1283 * Queries the engine for the set of features it supports for a given locale. 1284 * Features can either be framework defined, e.g. 1285 * {@link TextToSpeech.Engine#KEY_FEATURE_NETWORK_SYNTHESIS} or engine specific. 1286 * Engine specific keys must be prefixed by the name of the engine they 1287 * are intended for. These keys can be used as parameters to 1288 * {@link TextToSpeech#speak(String, int, java.util.HashMap)} and 1289 * {@link TextToSpeech#synthesizeToFile(String, java.util.HashMap, String)}. 1290 * 1291 * Features values are strings and their values must meet restrictions described in their 1292 * documentation. 1293 * 1294 * @param locale The locale to query features for. 1295 * @return Set instance. May return {@code null} on error. 1296 * @deprecated As of API level 21, please use voices. In order to query features of the voice, 1297 * call {@link #getVoices()} to retrieve the list of available voices and 1298 * {@link Voice#getFeatures()} to retrieve the set of features. 1299 */ 1300 @Deprecated 1301 public Set<String> getFeatures(final Locale locale) { 1302 return runAction(new Action<Set<String>>() { 1303 @Override 1304 public Set<String> run(ITextToSpeechService service) throws RemoteException { 1305 String[] features = null; 1306 try { 1307 features = service.getFeaturesForLanguage( 1308 locale.getISO3Language(), locale.getISO3Country(), locale.getVariant()); 1309 } catch(MissingResourceException e) { 1310 Log.w(TAG, "Couldn't retrieve 3 letter ISO 639-2/T language and/or ISO 3166 " + 1311 "country code for locale: " + locale, e); 1312 return null; 1313 } 1314 1315 if (features != null) { 1316 final Set<String> featureSet = new HashSet<String>(); 1317 Collections.addAll(featureSet, features); 1318 return featureSet; 1319 } 1320 return null; 1321 } 1322 }, null, "getFeatures"); 1323 } 1324 1325 /** 1326 * Checks whether the TTS engine is busy speaking. Note that a speech item is 1327 * considered complete once it's audio data has been sent to the audio mixer, or 1328 * written to a file. There might be a finite lag between this point, and when 1329 * the audio hardware completes playback. 1330 * 1331 * @return {@code true} if the TTS engine is speaking. 1332 */ 1333 public boolean isSpeaking() { 1334 return runAction(new Action<Boolean>() { 1335 @Override 1336 public Boolean run(ITextToSpeechService service) throws RemoteException { 1337 return service.isSpeaking(); 1338 } 1339 }, false, "isSpeaking"); 1340 } 1341 1342 /** 1343 * Interrupts the current utterance (whether played or rendered to file) and discards other 1344 * utterances in the queue. 1345 * 1346 * @return {@link #ERROR} or {@link #SUCCESS}. 1347 */ 1348 public int stop() { 1349 return runAction(new Action<Integer>() { 1350 @Override 1351 public Integer run(ITextToSpeechService service) throws RemoteException { 1352 return service.stop(getCallerIdentity()); 1353 } 1354 }, ERROR, "stop"); 1355 } 1356 1357 /** 1358 * Sets the speech rate. 1359 * 1360 * This has no effect on any pre-recorded speech. 1361 * 1362 * @param speechRate Speech rate. {@code 1.0} is the normal speech rate, 1363 * lower values slow down the speech ({@code 0.5} is half the normal speech rate), 1364 * greater values accelerate it ({@code 2.0} is twice the normal speech rate). 1365 * 1366 * @return {@link #ERROR} or {@link #SUCCESS}. 1367 */ 1368 public int setSpeechRate(float speechRate) { 1369 if (speechRate > 0.0f) { 1370 int intRate = (int)(speechRate * 100); 1371 if (intRate > 0) { 1372 synchronized (mStartLock) { 1373 mParams.putInt(Engine.KEY_PARAM_RATE, intRate); 1374 } 1375 return SUCCESS; 1376 } 1377 } 1378 return ERROR; 1379 } 1380 1381 /** 1382 * Sets the speech pitch for the TextToSpeech engine. 1383 * 1384 * This has no effect on any pre-recorded speech. 1385 * 1386 * @param pitch Speech pitch. {@code 1.0} is the normal pitch, 1387 * lower values lower the tone of the synthesized voice, 1388 * greater values increase it. 1389 * 1390 * @return {@link #ERROR} or {@link #SUCCESS}. 1391 */ 1392 public int setPitch(float pitch) { 1393 if (pitch > 0.0f) { 1394 int intPitch = (int)(pitch * 100); 1395 if (intPitch > 0) { 1396 synchronized (mStartLock) { 1397 mParams.putInt(Engine.KEY_PARAM_PITCH, intPitch); 1398 } 1399 return SUCCESS; 1400 } 1401 } 1402 return ERROR; 1403 } 1404 1405 /** 1406 * Sets the audio attributes to be used when speaking text or playing 1407 * back a file. 1408 * 1409 * @param audioAttributes Valid AudioAttributes instance. 1410 * 1411 * @return {@link #ERROR} or {@link #SUCCESS}. 1412 */ 1413 public int setAudioAttributes(AudioAttributes audioAttributes) { 1414 if (audioAttributes != null) { 1415 synchronized (mStartLock) { 1416 mParams.putParcelable(Engine.KEY_PARAM_AUDIO_ATTRIBUTES, 1417 audioAttributes); 1418 } 1419 return SUCCESS; 1420 } 1421 return ERROR; 1422 } 1423 1424 /** 1425 * @return the engine currently in use by this TextToSpeech instance. 1426 * @hide 1427 */ 1428 public String getCurrentEngine() { 1429 return mCurrentEngine; 1430 } 1431 1432 /** 1433 * Returns a Locale instance describing the language currently being used as the default 1434 * Text-to-speech language. 1435 * 1436 * The locale object returned by this method is NOT a valid one. It has identical form to the 1437 * one in {@link #getLanguage()}. Please refer to {@link #getLanguage()} for more information. 1438 * 1439 * @return language, country (if any) and variant (if any) used by the client stored in a 1440 * Locale instance, or {@code null} on error. 1441 * @deprecated As of API level 21, use <code>getDefaultVoice().getLocale()</code> ({@link 1442 * #getDefaultVoice()}) 1443 */ 1444 @Deprecated 1445 public Locale getDefaultLanguage() { 1446 return runAction(new Action<Locale>() { 1447 @Override 1448 public Locale run(ITextToSpeechService service) throws RemoteException { 1449 String[] defaultLanguage = service.getClientDefaultLanguage(); 1450 1451 return new Locale(defaultLanguage[0], defaultLanguage[1], defaultLanguage[2]); 1452 } 1453 }, null, "getDefaultLanguage"); 1454 } 1455 1456 /** 1457 * Sets the text-to-speech language. 1458 * The TTS engine will try to use the closest match to the specified 1459 * language as represented by the Locale, but there is no guarantee that the exact same Locale 1460 * will be used. Use {@link #isLanguageAvailable(Locale)} to check the level of support 1461 * before choosing the language to use for the next utterances. 1462 * 1463 * This method sets the current voice to the default one for the given Locale; 1464 * {@link #getVoice()} can be used to retrieve it. 1465 * 1466 * @param loc The locale describing the language to be used. 1467 * 1468 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1469 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1470 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1471 */ 1472 public int setLanguage(final Locale loc) { 1473 return runAction(new Action<Integer>() { 1474 @Override 1475 public Integer run(ITextToSpeechService service) throws RemoteException { 1476 if (loc == null) { 1477 return LANG_NOT_SUPPORTED; 1478 } 1479 String language = null, country = null; 1480 try { 1481 language = loc.getISO3Language(); 1482 } catch (MissingResourceException e) { 1483 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1484 return LANG_NOT_SUPPORTED; 1485 } 1486 1487 try { 1488 country = loc.getISO3Country(); 1489 } catch (MissingResourceException e) { 1490 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1491 return LANG_NOT_SUPPORTED; 1492 } 1493 1494 String variant = loc.getVariant(); 1495 1496 // As of API level 21, setLanguage is implemented using setVoice. 1497 // (which, in the default implementation, will call loadLanguage on the service 1498 // interface). 1499 1500 // Sanitize locale using isLanguageAvailable. 1501 int result = service.isLanguageAvailable(language, country, variant); 1502 if (result >= LANG_AVAILABLE) { 1503 // Get the default voice for the locale. 1504 String voiceName = service.getDefaultVoiceNameFor(language, country, variant); 1505 if (TextUtils.isEmpty(voiceName)) { 1506 Log.w(TAG, "Couldn't find the default voice for " + language + "-" + 1507 country + "-" + variant); 1508 return LANG_NOT_SUPPORTED; 1509 } 1510 1511 // Load it. 1512 if (service.loadVoice(getCallerIdentity(), voiceName) == TextToSpeech.ERROR) { 1513 Log.w(TAG, "The service claimed " + language + "-" + country + "-" 1514 + variant + " was available with voice name " + voiceName 1515 + " but loadVoice returned ERROR"); 1516 return LANG_NOT_SUPPORTED; 1517 } 1518 1519 // Set the language/country/variant of the voice, so #getLanguage will return 1520 // the currently set voice locale when called. 1521 Voice voice = getVoice(service, voiceName); 1522 if (voice == null) { 1523 Log.w(TAG, "getDefaultVoiceNameFor returned " + voiceName + " for locale " 1524 + language + "-" + country + "-" + variant 1525 + " but getVoice returns null"); 1526 return LANG_NOT_SUPPORTED; 1527 } 1528 String voiceLanguage = ""; 1529 try { 1530 voiceLanguage = voice.getLocale().getISO3Language(); 1531 } catch (MissingResourceException e) { 1532 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + 1533 voice.getLocale(), e); 1534 } 1535 1536 String voiceCountry = ""; 1537 try { 1538 voiceCountry = voice.getLocale().getISO3Country(); 1539 } catch (MissingResourceException e) { 1540 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + 1541 voice.getLocale(), e); 1542 } 1543 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, voiceName); 1544 mParams.putString(Engine.KEY_PARAM_LANGUAGE, voiceLanguage); 1545 mParams.putString(Engine.KEY_PARAM_COUNTRY, voiceCountry); 1546 mParams.putString(Engine.KEY_PARAM_VARIANT, voice.getLocale().getVariant()); 1547 } 1548 return result; 1549 } 1550 }, LANG_NOT_SUPPORTED, "setLanguage"); 1551 } 1552 1553 /** 1554 * Returns a Locale instance describing the language currently being used for synthesis 1555 * requests sent to the TextToSpeech engine. 1556 * 1557 * In Android 4.2 and before (API <= 17) this function returns the language that is currently 1558 * being used by the TTS engine. That is the last language set by this or any other 1559 * client by a {@link TextToSpeech#setLanguage} call to the same engine. 1560 * 1561 * In Android versions after 4.2 this function returns the language that is currently being 1562 * used for the synthesis requests sent from this client. That is the last language set 1563 * by a {@link TextToSpeech#setLanguage} call on this instance. 1564 * 1565 * If a voice is set (by {@link #setVoice(Voice)}), getLanguage will return the language of 1566 * the currently set voice. 1567 * 1568 * Please note that the Locale object returned by this method is NOT a valid Locale object. Its 1569 * language field contains a three-letter ISO 639-2/T code (where a proper Locale would use 1570 * a two-letter ISO 639-1 code), and the country field contains a three-letter ISO 3166 country 1571 * code (where a proper Locale would use a two-letter ISO 3166-1 code). 1572 * 1573 * @return language, country (if any) and variant (if any) used by the client stored in a 1574 * Locale instance, or {@code null} on error. 1575 * 1576 * @deprecated As of API level 21, please use <code>getVoice().getLocale()</code> 1577 * ({@link #getVoice()}). 1578 */ 1579 @Deprecated 1580 public Locale getLanguage() { 1581 return runAction(new Action<Locale>() { 1582 @Override 1583 public Locale run(ITextToSpeechService service) { 1584 /* No service call, but we're accessing mParams, hence need for 1585 wrapping it as an Action instance */ 1586 String lang = mParams.getString(Engine.KEY_PARAM_LANGUAGE, ""); 1587 String country = mParams.getString(Engine.KEY_PARAM_COUNTRY, ""); 1588 String variant = mParams.getString(Engine.KEY_PARAM_VARIANT, ""); 1589 return new Locale(lang, country, variant); 1590 } 1591 }, null, "getLanguage"); 1592 } 1593 1594 /** 1595 * Query the engine about the set of available languages. 1596 */ 1597 public Set<Locale> getAvailableLanguages() { 1598 return runAction(new Action<Set<Locale>>() { 1599 @Override 1600 public Set<Locale> run(ITextToSpeechService service) throws RemoteException { 1601 List<Voice> voices = service.getVoices(); 1602 if (voices == null) { 1603 return new HashSet<Locale>(); 1604 } 1605 HashSet<Locale> locales = new HashSet<Locale>(); 1606 for (Voice voice : voices) { 1607 locales.add(voice.getLocale()); 1608 } 1609 return locales; 1610 } 1611 }, null, "getAvailableLanguages"); 1612 } 1613 1614 /** 1615 * Query the engine about the set of available voices. 1616 * 1617 * Each TTS Engine can expose multiple voices for each locale, each with a different set of 1618 * features. 1619 * 1620 * @see #setVoice(Voice) 1621 * @see Voice 1622 */ 1623 public Set<Voice> getVoices() { 1624 return runAction(new Action<Set<Voice>>() { 1625 @Override 1626 public Set<Voice> run(ITextToSpeechService service) throws RemoteException { 1627 List<Voice> voices = service.getVoices(); 1628 return (voices != null) ? new HashSet<Voice>(voices) : new HashSet<Voice>(); 1629 } 1630 }, null, "getVoices"); 1631 } 1632 1633 /** 1634 * Sets the text-to-speech voice. 1635 * 1636 * @param voice One of objects returned by {@link #getVoices()}. 1637 * 1638 * @return {@link #ERROR} or {@link #SUCCESS}. 1639 * 1640 * @see #getVoices 1641 * @see Voice 1642 */ 1643 public int setVoice(final Voice voice) { 1644 return runAction(new Action<Integer>() { 1645 @Override 1646 public Integer run(ITextToSpeechService service) throws RemoteException { 1647 int result = service.loadVoice(getCallerIdentity(), voice.getName()); 1648 if (result == SUCCESS) { 1649 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, voice.getName()); 1650 1651 // Set the language/country/variant, so #getLanguage will return the voice 1652 // locale when called. 1653 String language = ""; 1654 try { 1655 language = voice.getLocale().getISO3Language(); 1656 } catch (MissingResourceException e) { 1657 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + 1658 voice.getLocale(), e); 1659 } 1660 1661 String country = ""; 1662 try { 1663 country = voice.getLocale().getISO3Country(); 1664 } catch (MissingResourceException e) { 1665 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + 1666 voice.getLocale(), e); 1667 } 1668 mParams.putString(Engine.KEY_PARAM_LANGUAGE, language); 1669 mParams.putString(Engine.KEY_PARAM_COUNTRY, country); 1670 mParams.putString(Engine.KEY_PARAM_VARIANT, voice.getLocale().getVariant()); 1671 } 1672 return result; 1673 } 1674 }, LANG_NOT_SUPPORTED, "setVoice"); 1675 } 1676 1677 /** 1678 * Returns a Voice instance describing the voice currently being used for synthesis 1679 * requests sent to the TextToSpeech engine. 1680 * 1681 * @return Voice instance used by the client, or {@code null} if not set or on error. 1682 * 1683 * @see #getVoices 1684 * @see #setVoice 1685 * @see Voice 1686 */ 1687 public Voice getVoice() { 1688 return runAction(new Action<Voice>() { 1689 @Override 1690 public Voice run(ITextToSpeechService service) throws RemoteException { 1691 String voiceName = mParams.getString(Engine.KEY_PARAM_VOICE_NAME, ""); 1692 if (TextUtils.isEmpty(voiceName)) { 1693 return null; 1694 } 1695 return getVoice(service, voiceName); 1696 } 1697 }, null, "getVoice"); 1698 } 1699 1700 1701 /** 1702 * Returns a Voice instance of the voice with the given voice name. 1703 * 1704 * @return Voice instance with the given voice name, or {@code null} if not set or on error. 1705 * 1706 * @see Voice 1707 */ 1708 private Voice getVoice(ITextToSpeechService service, String voiceName) throws RemoteException { 1709 List<Voice> voices = service.getVoices(); 1710 if (voices == null) { 1711 Log.w(TAG, "getVoices returned null"); 1712 return null; 1713 } 1714 for (Voice voice : voices) { 1715 if (voice.getName().equals(voiceName)) { 1716 return voice; 1717 } 1718 } 1719 Log.w(TAG, "Could not find voice " + voiceName + " in voice list"); 1720 return null; 1721 } 1722 1723 /** 1724 * Returns a Voice instance that's the default voice for the default Text-to-speech language. 1725 * @return The default voice instance for the default language, or {@code null} if not set or 1726 * on error. 1727 */ 1728 public Voice getDefaultVoice() { 1729 return runAction(new Action<Voice>() { 1730 @Override 1731 public Voice run(ITextToSpeechService service) throws RemoteException { 1732 1733 String[] defaultLanguage = service.getClientDefaultLanguage(); 1734 1735 if (defaultLanguage == null || defaultLanguage.length == 0) { 1736 Log.e(TAG, "service.getClientDefaultLanguage() returned empty array"); 1737 return null; 1738 } 1739 String language = defaultLanguage[0]; 1740 String country = (defaultLanguage.length > 1) ? defaultLanguage[1] : ""; 1741 String variant = (defaultLanguage.length > 2) ? defaultLanguage[2] : ""; 1742 1743 // Sanitize the locale using isLanguageAvailable. 1744 int result = service.isLanguageAvailable(language, country, variant); 1745 if (result < LANG_AVAILABLE) { 1746 // The default language is not supported. 1747 return null; 1748 } 1749 1750 // Get the default voice name 1751 String voiceName = service.getDefaultVoiceNameFor(language, country, variant); 1752 if (TextUtils.isEmpty(voiceName)) { 1753 return null; 1754 } 1755 1756 // Find it 1757 List<Voice> voices = service.getVoices(); 1758 if (voices == null) { 1759 return null; 1760 } 1761 for (Voice voice : voices) { 1762 if (voice.getName().equals(voiceName)) { 1763 return voice; 1764 } 1765 } 1766 return null; 1767 } 1768 }, null, "getDefaultVoice"); 1769 } 1770 1771 1772 1773 /** 1774 * Checks if the specified language as represented by the Locale is available and supported. 1775 * 1776 * @param loc The Locale describing the language to be used. 1777 * 1778 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1779 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1780 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1781 */ 1782 public int isLanguageAvailable(final Locale loc) { 1783 return runAction(new Action<Integer>() { 1784 @Override 1785 public Integer run(ITextToSpeechService service) throws RemoteException { 1786 String language = null, country = null; 1787 1788 try { 1789 language = loc.getISO3Language(); 1790 } catch (MissingResourceException e) { 1791 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1792 return LANG_NOT_SUPPORTED; 1793 } 1794 1795 try { 1796 country = loc.getISO3Country(); 1797 } catch (MissingResourceException e) { 1798 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1799 return LANG_NOT_SUPPORTED; 1800 } 1801 1802 return service.isLanguageAvailable(language, country, loc.getVariant()); 1803 } 1804 }, LANG_NOT_SUPPORTED, "isLanguageAvailable"); 1805 } 1806 1807 /** 1808 * Synthesizes the given text to a file using the specified parameters. 1809 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1810 * requests and then returns. The synthesis might not have finished (or even started!) at the 1811 * time when this method returns. In order to reliably detect errors during synthesis, 1812 * we recommend setting an utterance progress listener (see 1813 * {@link #setOnUtteranceProgressListener}). 1814 * 1815 * @param text The text that should be synthesized. No longer than 1816 * {@link #getMaxSpeechInputLength()} characters. 1817 * @param params Parameters for the request. Can be null. 1818 * Engine specific parameters may be passed in but the parameter keys 1819 * must be prefixed by the name of the engine they are intended for. For example 1820 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1821 * engine named "com.svox.pico" if it is being used. 1822 * @param file File to write the generated audio data to. 1823 * @param utteranceId An unique identifier for this request. 1824 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the synthesizeToFile operation. 1825 */ 1826 public int synthesizeToFile(final CharSequence text, final Bundle params, 1827 final File file, final String utteranceId) { 1828 return runAction(new Action<Integer>() { 1829 @Override 1830 public Integer run(ITextToSpeechService service) throws RemoteException { 1831 ParcelFileDescriptor fileDescriptor; 1832 int returnValue; 1833 try { 1834 if(file.exists() && !file.canWrite()) { 1835 Log.e(TAG, "Can't write to " + file); 1836 return ERROR; 1837 } 1838 fileDescriptor = ParcelFileDescriptor.open(file, 1839 ParcelFileDescriptor.MODE_WRITE_ONLY | 1840 ParcelFileDescriptor.MODE_CREATE | 1841 ParcelFileDescriptor.MODE_TRUNCATE); 1842 returnValue = service.synthesizeToFileDescriptor(getCallerIdentity(), text, 1843 fileDescriptor, getParams(params), utteranceId); 1844 fileDescriptor.close(); 1845 return returnValue; 1846 } catch (FileNotFoundException e) { 1847 Log.e(TAG, "Opening file " + file + " failed", e); 1848 return ERROR; 1849 } catch (IOException e) { 1850 Log.e(TAG, "Closing file " + file + " failed", e); 1851 return ERROR; 1852 } 1853 } 1854 }, ERROR, "synthesizeToFile"); 1855 } 1856 1857 /** 1858 * Synthesizes the given text to a file using the specified parameters. 1859 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1860 * requests and then returns. The synthesis might not have finished (or even started!) at the 1861 * time when this method returns. In order to reliably detect errors during synthesis, 1862 * we recommend setting an utterance progress listener (see 1863 * {@link #setOnUtteranceProgressListener}) and using the 1864 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1865 * 1866 * @param text The text that should be synthesized. No longer than 1867 * {@link #getMaxSpeechInputLength()} characters. 1868 * @param params Parameters for the request. Can be null. 1869 * Supported parameter names: 1870 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1871 * Engine specific parameters may be passed in but the parameter keys 1872 * must be prefixed by the name of the engine they are intended for. For example 1873 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1874 * engine named "com.svox.pico" if it is being used. 1875 * @param filename Absolute file filename to write the generated audio data to.It should be 1876 * something like "/sdcard/myappsounds/mysound.wav". 1877 * 1878 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the synthesizeToFile operation. 1879 * @deprecated As of API level 21, replaced by 1880 * {@link #synthesizeToFile(CharSequence, Bundle, File, String)}. 1881 */ 1882 @Deprecated 1883 public int synthesizeToFile(final String text, final HashMap<String, String> params, 1884 final String filename) { 1885 return synthesizeToFile(text, convertParamsHashMaptoBundle(params), 1886 new File(filename), params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1887 } 1888 1889 private Bundle convertParamsHashMaptoBundle(HashMap<String, String> params) { 1890 if (params != null && !params.isEmpty()) { 1891 Bundle bundle = new Bundle(); 1892 copyIntParam(bundle, params, Engine.KEY_PARAM_STREAM); 1893 copyIntParam(bundle, params, Engine.KEY_PARAM_SESSION_ID); 1894 copyStringParam(bundle, params, Engine.KEY_PARAM_UTTERANCE_ID); 1895 copyFloatParam(bundle, params, Engine.KEY_PARAM_VOLUME); 1896 copyFloatParam(bundle, params, Engine.KEY_PARAM_PAN); 1897 1898 // Copy feature strings defined by the framework. 1899 copyStringParam(bundle, params, Engine.KEY_FEATURE_NETWORK_SYNTHESIS); 1900 copyStringParam(bundle, params, Engine.KEY_FEATURE_EMBEDDED_SYNTHESIS); 1901 copyIntParam(bundle, params, Engine.KEY_FEATURE_NETWORK_TIMEOUT_MS); 1902 copyIntParam(bundle, params, Engine.KEY_FEATURE_NETWORK_RETRIES_COUNT); 1903 1904 // Copy over all parameters that start with the name of the 1905 // engine that we are currently connected to. The engine is 1906 // free to interpret them as it chooses. 1907 if (!TextUtils.isEmpty(mCurrentEngine)) { 1908 for (Map.Entry<String, String> entry : params.entrySet()) { 1909 final String key = entry.getKey(); 1910 if (key != null && key.startsWith(mCurrentEngine)) { 1911 bundle.putString(key, entry.getValue()); 1912 } 1913 } 1914 } 1915 1916 return bundle; 1917 } 1918 return null; 1919 } 1920 1921 private Bundle getParams(Bundle params) { 1922 if (params != null && !params.isEmpty()) { 1923 Bundle bundle = new Bundle(mParams); 1924 bundle.putAll(params); 1925 1926 verifyIntegerBundleParam(bundle, Engine.KEY_PARAM_STREAM); 1927 verifyIntegerBundleParam(bundle, Engine.KEY_PARAM_SESSION_ID); 1928 verifyStringBundleParam(bundle, Engine.KEY_PARAM_UTTERANCE_ID); 1929 verifyFloatBundleParam(bundle, Engine.KEY_PARAM_VOLUME); 1930 verifyFloatBundleParam(bundle, Engine.KEY_PARAM_PAN); 1931 1932 // Copy feature strings defined by the framework. 1933 verifyBooleanBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_SYNTHESIS); 1934 verifyBooleanBundleParam(bundle, Engine.KEY_FEATURE_EMBEDDED_SYNTHESIS); 1935 verifyIntegerBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_TIMEOUT_MS); 1936 verifyIntegerBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_RETRIES_COUNT); 1937 1938 return bundle; 1939 } else { 1940 return mParams; 1941 } 1942 } 1943 1944 private static boolean verifyIntegerBundleParam(Bundle bundle, String key) { 1945 if (bundle.containsKey(key)) { 1946 if (!(bundle.get(key) instanceof Integer || 1947 bundle.get(key) instanceof Long)) { 1948 bundle.remove(key); 1949 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1950 + " with invalid type. Should be an Integer or a Long"); 1951 return false; 1952 } 1953 } 1954 return true; 1955 } 1956 1957 private static boolean verifyStringBundleParam(Bundle bundle, String key) { 1958 if (bundle.containsKey(key)) { 1959 if (!(bundle.get(key) instanceof String)) { 1960 bundle.remove(key); 1961 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1962 + " with invalid type. Should be a String"); 1963 return false; 1964 } 1965 } 1966 return true; 1967 } 1968 1969 private static boolean verifyBooleanBundleParam(Bundle bundle, String key) { 1970 if (bundle.containsKey(key)) { 1971 if (!(bundle.get(key) instanceof Boolean || 1972 bundle.get(key) instanceof String)) { 1973 bundle.remove(key); 1974 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1975 + " with invalid type. Should be a Boolean or String"); 1976 return false; 1977 } 1978 } 1979 return true; 1980 } 1981 1982 1983 private static boolean verifyFloatBundleParam(Bundle bundle, String key) { 1984 if (bundle.containsKey(key)) { 1985 if (!(bundle.get(key) instanceof Float || 1986 bundle.get(key) instanceof Double)) { 1987 bundle.remove(key); 1988 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1989 + " with invalid type. Should be a Float or a Double"); 1990 return false; 1991 } 1992 } 1993 return true; 1994 } 1995 1996 private void copyStringParam(Bundle bundle, HashMap<String, String> params, String key) { 1997 String value = params.get(key); 1998 if (value != null) { 1999 bundle.putString(key, value); 2000 } 2001 } 2002 2003 private void copyIntParam(Bundle bundle, HashMap<String, String> params, String key) { 2004 String valueString = params.get(key); 2005 if (!TextUtils.isEmpty(valueString)) { 2006 try { 2007 int value = Integer.parseInt(valueString); 2008 bundle.putInt(key, value); 2009 } catch (NumberFormatException ex) { 2010 // don't set the value in the bundle 2011 } 2012 } 2013 } 2014 2015 private void copyFloatParam(Bundle bundle, HashMap<String, String> params, String key) { 2016 String valueString = params.get(key); 2017 if (!TextUtils.isEmpty(valueString)) { 2018 try { 2019 float value = Float.parseFloat(valueString); 2020 bundle.putFloat(key, value); 2021 } catch (NumberFormatException ex) { 2022 // don't set the value in the bundle 2023 } 2024 } 2025 } 2026 2027 /** 2028 * Sets the listener that will be notified when synthesis of an utterance completes. 2029 * 2030 * @param listener The listener to use. 2031 * 2032 * @return {@link #ERROR} or {@link #SUCCESS}. 2033 * 2034 * @deprecated Use {@link #setOnUtteranceProgressListener(UtteranceProgressListener)} 2035 * instead. 2036 */ 2037 @Deprecated 2038 public int setOnUtteranceCompletedListener(final OnUtteranceCompletedListener listener) { 2039 mUtteranceProgressListener = UtteranceProgressListener.from(listener); 2040 return TextToSpeech.SUCCESS; 2041 } 2042 2043 /** 2044 * Sets the listener that will be notified of various events related to the 2045 * synthesis of a given utterance. 2046 * 2047 * See {@link UtteranceProgressListener} and 2048 * {@link TextToSpeech.Engine#KEY_PARAM_UTTERANCE_ID}. 2049 * 2050 * @param listener the listener to use. 2051 * @return {@link #ERROR} or {@link #SUCCESS} 2052 */ 2053 public int setOnUtteranceProgressListener(UtteranceProgressListener listener) { 2054 mUtteranceProgressListener = listener; 2055 return TextToSpeech.SUCCESS; 2056 } 2057 2058 /** 2059 * Sets the TTS engine to use. 2060 * 2061 * @deprecated This doesn't inform callers when the TTS engine has been 2062 * initialized. {@link #TextToSpeech(Context, OnInitListener, String)} 2063 * can be used with the appropriate engine name. Also, there is no 2064 * guarantee that the engine specified will be loaded. If it isn't 2065 * installed or disabled, the user / system wide defaults will apply. 2066 * 2067 * @param enginePackageName The package name for the synthesis engine (e.g. "com.svox.pico") 2068 * 2069 * @return {@link #ERROR} or {@link #SUCCESS}. 2070 */ 2071 @Deprecated 2072 public int setEngineByPackageName(String enginePackageName) { 2073 mRequestedEngine = enginePackageName; 2074 return initTts(); 2075 } 2076 2077 /** 2078 * Gets the package name of the default speech synthesis engine. 2079 * 2080 * @return Package name of the TTS engine that the user has chosen 2081 * as their default. 2082 */ 2083 public String getDefaultEngine() { 2084 return mEnginesHelper.getDefaultEngine(); 2085 } 2086 2087 /** 2088 * Checks whether the user's settings should override settings requested 2089 * by the calling application. As of the Ice cream sandwich release, 2090 * user settings never forcibly override the app's settings. 2091 */ 2092 @Deprecated 2093 public boolean areDefaultsEnforced() { 2094 return false; 2095 } 2096 2097 /** 2098 * Gets a list of all installed TTS engines. 2099 * 2100 * @return A list of engine info objects. The list can be empty, but never {@code null}. 2101 */ 2102 public List<EngineInfo> getEngines() { 2103 return mEnginesHelper.getEngines(); 2104 } 2105 2106 private class Connection implements ServiceConnection { 2107 private ITextToSpeechService mService; 2108 2109 private SetupConnectionAsyncTask mOnSetupConnectionAsyncTask; 2110 2111 private boolean mEstablished; 2112 2113 private final ITextToSpeechCallback.Stub mCallback = 2114 new ITextToSpeechCallback.Stub() { 2115 public void onStop(String utteranceId, boolean isStarted) 2116 throws RemoteException { 2117 UtteranceProgressListener listener = mUtteranceProgressListener; 2118 if (listener != null) { 2119 listener.onStop(utteranceId, isStarted); 2120 } 2121 }; 2122 2123 @Override 2124 public void onSuccess(String utteranceId) { 2125 UtteranceProgressListener listener = mUtteranceProgressListener; 2126 if (listener != null) { 2127 listener.onDone(utteranceId); 2128 } 2129 } 2130 2131 @Override 2132 public void onError(String utteranceId, int errorCode) { 2133 UtteranceProgressListener listener = mUtteranceProgressListener; 2134 if (listener != null) { 2135 listener.onError(utteranceId); 2136 } 2137 } 2138 2139 @Override 2140 public void onStart(String utteranceId) { 2141 UtteranceProgressListener listener = mUtteranceProgressListener; 2142 if (listener != null) { 2143 listener.onStart(utteranceId); 2144 } 2145 } 2146 2147 @Override 2148 public void onBeginSynthesis( 2149 String utteranceId, 2150 int sampleRateInHz, 2151 int audioFormat, 2152 int channelCount) { 2153 UtteranceProgressListener listener = mUtteranceProgressListener; 2154 if (listener != null) { 2155 listener.onBeginSynthesis( 2156 utteranceId, sampleRateInHz, audioFormat, channelCount); 2157 } 2158 } 2159 2160 @Override 2161 public void onAudioAvailable(String utteranceId, byte[] audio) { 2162 UtteranceProgressListener listener = mUtteranceProgressListener; 2163 if (listener != null) { 2164 listener.onAudioAvailable(utteranceId, audio); 2165 } 2166 } 2167 2168 @Override 2169 public void onRangeStart(String utteranceId, int start, int end, int frame) { 2170 UtteranceProgressListener listener = mUtteranceProgressListener; 2171 if (listener != null) { 2172 listener.onRangeStart(utteranceId, start, end, frame); 2173 } 2174 } 2175 }; 2176 2177 private class SetupConnectionAsyncTask extends AsyncTask<Void, Void, Integer> { 2178 private final ComponentName mName; 2179 2180 public SetupConnectionAsyncTask(ComponentName name) { 2181 mName = name; 2182 } 2183 2184 @Override 2185 protected Integer doInBackground(Void... params) { 2186 synchronized(mStartLock) { 2187 if (isCancelled()) { 2188 return null; 2189 } 2190 2191 try { 2192 mService.setCallback(getCallerIdentity(), mCallback); 2193 2194 if (mParams.getString(Engine.KEY_PARAM_LANGUAGE) == null) { 2195 String[] defaultLanguage = mService.getClientDefaultLanguage(); 2196 mParams.putString(Engine.KEY_PARAM_LANGUAGE, defaultLanguage[0]); 2197 mParams.putString(Engine.KEY_PARAM_COUNTRY, defaultLanguage[1]); 2198 mParams.putString(Engine.KEY_PARAM_VARIANT, defaultLanguage[2]); 2199 2200 // Get the default voice for the locale. 2201 String defaultVoiceName = mService.getDefaultVoiceNameFor( 2202 defaultLanguage[0], defaultLanguage[1], defaultLanguage[2]); 2203 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, defaultVoiceName); 2204 } 2205 2206 Log.i(TAG, "Set up connection to " + mName); 2207 return SUCCESS; 2208 } catch (RemoteException re) { 2209 Log.e(TAG, "Error connecting to service, setCallback() failed"); 2210 return ERROR; 2211 } 2212 } 2213 } 2214 2215 @Override 2216 protected void onPostExecute(Integer result) { 2217 synchronized(mStartLock) { 2218 if (mOnSetupConnectionAsyncTask == this) { 2219 mOnSetupConnectionAsyncTask = null; 2220 } 2221 mEstablished = true; 2222 dispatchOnInit(result); 2223 } 2224 } 2225 } 2226 2227 @Override 2228 public void onServiceConnected(ComponentName name, IBinder service) { 2229 synchronized(mStartLock) { 2230 mConnectingServiceConnection = null; 2231 2232 Log.i(TAG, "Connected to " + name); 2233 2234 if (mOnSetupConnectionAsyncTask != null) { 2235 mOnSetupConnectionAsyncTask.cancel(false); 2236 } 2237 2238 mService = ITextToSpeechService.Stub.asInterface(service); 2239 mServiceConnection = Connection.this; 2240 2241 mEstablished = false; 2242 mOnSetupConnectionAsyncTask = new SetupConnectionAsyncTask(name); 2243 mOnSetupConnectionAsyncTask.execute(); 2244 } 2245 } 2246 2247 public IBinder getCallerIdentity() { 2248 return mCallback; 2249 } 2250 2251 /** 2252 * Clear connection related fields and cancel mOnServiceConnectedAsyncTask if set. 2253 * 2254 * @return true if we cancel mOnSetupConnectionAsyncTask in progress. 2255 */ 2256 private boolean clearServiceConnection() { 2257 synchronized(mStartLock) { 2258 boolean result = false; 2259 if (mOnSetupConnectionAsyncTask != null) { 2260 result = mOnSetupConnectionAsyncTask.cancel(false); 2261 mOnSetupConnectionAsyncTask = null; 2262 } 2263 2264 mService = null; 2265 // If this is the active connection, clear it 2266 if (mServiceConnection == this) { 2267 mServiceConnection = null; 2268 } 2269 return result; 2270 } 2271 } 2272 2273 @Override 2274 public void onServiceDisconnected(ComponentName name) { 2275 Log.i(TAG, "Asked to disconnect from " + name); 2276 if (clearServiceConnection()) { 2277 /* We need to protect against a rare case where engine 2278 * dies just after successful connection - and we process onServiceDisconnected 2279 * before OnServiceConnectedAsyncTask.onPostExecute. onServiceDisconnected cancels 2280 * OnServiceConnectedAsyncTask.onPostExecute and we don't call dispatchOnInit 2281 * with ERROR as argument. 2282 */ 2283 dispatchOnInit(ERROR); 2284 } 2285 } 2286 2287 public void disconnect() { 2288 mContext.unbindService(this); 2289 clearServiceConnection(); 2290 } 2291 2292 public boolean isEstablished() { 2293 return mService != null && mEstablished; 2294 } 2295 2296 public <R> R runAction(Action<R> action, R errorResult, String method, 2297 boolean reconnect, boolean onlyEstablishedConnection) { 2298 synchronized (mStartLock) { 2299 try { 2300 if (mService == null) { 2301 Log.w(TAG, method + " failed: not connected to TTS engine"); 2302 return errorResult; 2303 } 2304 if (onlyEstablishedConnection && !isEstablished()) { 2305 Log.w(TAG, method + " failed: TTS engine connection not fully set up"); 2306 return errorResult; 2307 } 2308 return action.run(mService); 2309 } catch (RemoteException ex) { 2310 Log.e(TAG, method + " failed", ex); 2311 if (reconnect) { 2312 disconnect(); 2313 initTts(); 2314 } 2315 return errorResult; 2316 } 2317 } 2318 } 2319 } 2320 2321 private interface Action<R> { 2322 R run(ITextToSpeechService service) throws RemoteException; 2323 } 2324 2325 /** 2326 * Information about an installed text-to-speech engine. 2327 * 2328 * @see TextToSpeech#getEngines 2329 */ 2330 public static class EngineInfo { 2331 /** 2332 * Engine package name.. 2333 */ 2334 public String name; 2335 /** 2336 * Localized label for the engine. 2337 */ 2338 public String label; 2339 /** 2340 * Icon for the engine. 2341 */ 2342 public int icon; 2343 /** 2344 * Whether this engine is a part of the system 2345 * image. 2346 * 2347 * @hide 2348 */ 2349 public boolean system; 2350 /** 2351 * The priority the engine declares for the the intent filter 2352 * {@code android.intent.action.TTS_SERVICE} 2353 * 2354 * @hide 2355 */ 2356 public int priority; 2357 2358 @Override 2359 public String toString() { 2360 return "EngineInfo{name=" + name + "}"; 2361 } 2362 2363 } 2364 2365 /** 2366 * Limit of length of input string passed to speak and synthesizeToFile. 2367 * 2368 * @see #speak 2369 * @see #synthesizeToFile 2370 */ 2371 public static int getMaxSpeechInputLength() { 2372 return 4000; 2373 } 2374 } 2375