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.annotation.UnsupportedAppUsage; 24 import android.content.ComponentName; 25 import android.content.ContentResolver; 26 import android.content.Context; 27 import android.content.Intent; 28 import android.content.ServiceConnection; 29 import android.media.AudioAttributes; 30 import android.media.AudioManager; 31 import android.net.Uri; 32 import android.os.AsyncTask; 33 import android.os.Bundle; 34 import android.os.IBinder; 35 import android.os.ParcelFileDescriptor; 36 import android.os.RemoteException; 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 is interested in information about what 491 * changed, it should use the ACTION_CHECK_TTS_DATA 492 * intent to discover available voices. 493 */ 494 @Deprecated 495 public static final String EXTRA_TTS_DATA_INSTALLED = "dataInstalled"; 496 497 // keys for the parameters passed with speak commands. Hidden keys are used internally 498 // to maintain engine state for each TextToSpeech instance. 499 /** 500 * @hide 501 */ 502 public static final String KEY_PARAM_RATE = "rate"; 503 504 /** 505 * @hide 506 */ 507 public static final String KEY_PARAM_VOICE_NAME = "voiceName"; 508 509 /** 510 * @hide 511 */ 512 public static final String KEY_PARAM_LANGUAGE = "language"; 513 514 /** 515 * @hide 516 */ 517 public static final String KEY_PARAM_COUNTRY = "country"; 518 519 /** 520 * @hide 521 */ 522 public static final String KEY_PARAM_VARIANT = "variant"; 523 524 /** 525 * @hide 526 */ 527 public static final String KEY_PARAM_ENGINE = "engine"; 528 529 /** 530 * @hide 531 */ 532 public static final String KEY_PARAM_PITCH = "pitch"; 533 534 /** 535 * Parameter key to specify the audio stream type to be used when speaking text 536 * or playing back a file. The value should be one of the STREAM_ constants 537 * defined in {@link AudioManager}. 538 * 539 * @see TextToSpeech#speak(CharSequence, int, Bundle, String) 540 * @see TextToSpeech#playEarcon(String, int, HashMap) 541 */ 542 public static final String KEY_PARAM_STREAM = "streamType"; 543 544 /** 545 * Parameter key to specify the audio attributes to be used when 546 * speaking text or playing back a file. The value should be set 547 * using {@link TextToSpeech#setAudioAttributes(AudioAttributes)}. 548 * 549 * @see TextToSpeech#speak(CharSequence, int, Bundle, String) 550 * @see TextToSpeech#playEarcon(String, int, HashMap) 551 * @hide 552 */ 553 public static final String KEY_PARAM_AUDIO_ATTRIBUTES = "audioAttributes"; 554 555 /** 556 * Parameter key to identify an utterance in the 557 * {@link TextToSpeech.OnUtteranceCompletedListener} after text has been 558 * spoken, a file has been played back or a silence duration has elapsed. 559 * 560 * @see TextToSpeech#speak(CharSequence, int, Bundle, String) 561 * @see TextToSpeech#playEarcon(String, int, HashMap) 562 * @see TextToSpeech#synthesizeToFile(String, HashMap, String) 563 */ 564 public static final String KEY_PARAM_UTTERANCE_ID = "utteranceId"; 565 566 /** 567 * Parameter key to specify the speech volume relative to the current stream type 568 * volume used when speaking text. Volume is specified as a float ranging from 0 to 1 569 * where 0 is silence, and 1 is the maximum volume (the default behavior). 570 * 571 * @see TextToSpeech#speak(CharSequence, int, Bundle, String) 572 * @see TextToSpeech#playEarcon(String, int, HashMap) 573 */ 574 public static final String KEY_PARAM_VOLUME = "volume"; 575 576 /** 577 * Parameter key to specify how the speech is panned from left to right when speaking text. 578 * Pan is specified as a float ranging from -1 to +1 where -1 maps to a hard-left pan, 579 * 0 to center (the default behavior), and +1 to hard-right. 580 * 581 * @see TextToSpeech#speak(CharSequence, int, Bundle, String) 582 * @see TextToSpeech#playEarcon(String, int, HashMap) 583 */ 584 public static final String KEY_PARAM_PAN = "pan"; 585 586 /** 587 * Feature key for network synthesis. See {@link TextToSpeech#getFeatures(Locale)} 588 * for a description of how feature keys work. If set (and supported by the engine 589 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must 590 * use network based synthesis. 591 * 592 * @see TextToSpeech#speak(CharSequence, int, Bundle, String) 593 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 594 * @see TextToSpeech#getFeatures(java.util.Locale) 595 * 596 * @deprecated Starting from API level 21, to select network synthesis, call 597 * {@link TextToSpeech#getVoices()}, find a suitable network voice 598 * ({@link Voice#isNetworkConnectionRequired()}) and pass it 599 * to {@link TextToSpeech#setVoice(Voice)}. 600 */ 601 @Deprecated 602 public static final String KEY_FEATURE_NETWORK_SYNTHESIS = "networkTts"; 603 604 /** 605 * Feature key for embedded synthesis. See {@link TextToSpeech#getFeatures(Locale)} 606 * for a description of how feature keys work. If set and supported by the engine 607 * as per {@link TextToSpeech#getFeatures(Locale)}, the engine must synthesize 608 * text on-device (without making network requests). 609 * 610 * @see TextToSpeech#speak(CharSequence, int, Bundle, String) 611 * @see TextToSpeech#synthesizeToFile(String, java.util.HashMap, String) 612 * @see TextToSpeech#getFeatures(java.util.Locale) 613 614 * @deprecated Starting from API level 21, to select embedded synthesis, call 615 * ({@link TextToSpeech#getVoices()}, find a suitable embedded voice 616 * ({@link Voice#isNetworkConnectionRequired()}) and pass it 617 * to {@link TextToSpeech#setVoice(Voice)}). 618 */ 619 @Deprecated 620 public static final String KEY_FEATURE_EMBEDDED_SYNTHESIS = "embeddedTts"; 621 622 /** 623 * Parameter key to specify an audio session identifier (obtained from 624 * {@link AudioManager#generateAudioSessionId()}) that will be used by the request audio 625 * output. It can be used to associate one of the {@link android.media.audiofx.AudioEffect} 626 * objects with the synthesis (or earcon) output. 627 * 628 * @see TextToSpeech#speak(CharSequence, int, Bundle, String) 629 * @see TextToSpeech#playEarcon(String, int, HashMap) 630 */ 631 public static final String KEY_PARAM_SESSION_ID = "sessionId"; 632 633 /** 634 * Feature key that indicates that the voice may need to download additional data to be fully 635 * functional. The download will be triggered by calling 636 * {@link TextToSpeech#setVoice(Voice)} or {@link TextToSpeech#setLanguage(Locale)}. 637 * Until download is complete, each synthesis request will either report 638 * {@link TextToSpeech#ERROR_NOT_INSTALLED_YET} error, or use a different voice to synthesize 639 * the request. This feature should NOT be used as a key of a request parameter. 640 * 641 * @see TextToSpeech#getFeatures(java.util.Locale) 642 * @see Voice#getFeatures() 643 */ 644 public static final String KEY_FEATURE_NOT_INSTALLED = "notInstalled"; 645 646 /** 647 * Feature key that indicate that a network timeout can be set for the request. If set and 648 * supported as per {@link TextToSpeech#getFeatures(Locale)} or {@link Voice#getFeatures()}, 649 * it can be used as request parameter to set the maximum allowed time for a single 650 * request attempt, in milliseconds, before synthesis fails. When used as a key of 651 * a request parameter, its value should be a string with an integer value. 652 * 653 * @see TextToSpeech#getFeatures(java.util.Locale) 654 * @see Voice#getFeatures() 655 */ 656 public static final String KEY_FEATURE_NETWORK_TIMEOUT_MS = "networkTimeoutMs"; 657 658 /** 659 * Feature key that indicates that network request retries count can be set for the request. 660 * If set and supported as per {@link TextToSpeech#getFeatures(Locale)} or 661 * {@link Voice#getFeatures()}, it can be used as a request parameter to set the 662 * number of network request retries that are attempted in case of failure. When used as 663 * a key of a request parameter, its value should be a string with an integer value. 664 * 665 * @see TextToSpeech#getFeatures(java.util.Locale) 666 * @see Voice#getFeatures() 667 */ 668 public static final String KEY_FEATURE_NETWORK_RETRIES_COUNT = "networkRetriesCount"; 669 } 670 671 private final Context mContext; 672 @UnsupportedAppUsage 673 private Connection mConnectingServiceConnection; 674 private Connection mServiceConnection; 675 @UnsupportedAppUsage 676 private OnInitListener mInitListener; 677 // Written from an unspecified application thread, read from 678 // a binder thread. 679 @Nullable private volatile UtteranceProgressListener mUtteranceProgressListener; 680 private final Object mStartLock = new Object(); 681 682 private String mRequestedEngine; 683 // Whether to initialize this TTS object with the default engine, 684 // if the requested engine is not available. Valid only if mRequestedEngine 685 // is not null. Used only for testing, though potentially useful API wise 686 // too. 687 private final boolean mUseFallback; 688 private final Map<String, Uri> mEarcons; 689 private final Map<CharSequence, Uri> mUtterances; 690 private final Bundle mParams = new Bundle(); 691 private final TtsEngines mEnginesHelper; 692 @UnsupportedAppUsage 693 private volatile String mCurrentEngine = null; 694 695 /** 696 * The constructor for the TextToSpeech class, using the default TTS engine. 697 * This will also initialize the associated TextToSpeech engine if it isn't already running. 698 * 699 * @param context 700 * The context this instance is running in. 701 * @param listener 702 * The {@link TextToSpeech.OnInitListener} that will be called when the 703 * TextToSpeech engine has initialized. In a case of a failure the listener 704 * may be called immediately, before TextToSpeech instance is fully constructed. 705 */ TextToSpeech(Context context, OnInitListener listener)706 public TextToSpeech(Context context, OnInitListener listener) { 707 this(context, listener, null); 708 } 709 710 /** 711 * The constructor for the TextToSpeech class, using the given TTS engine. 712 * This will also initialize the associated TextToSpeech engine if it isn't already running. 713 * 714 * @param context 715 * The context this instance is running in. 716 * @param listener 717 * The {@link TextToSpeech.OnInitListener} that will be called when the 718 * TextToSpeech engine has initialized. In a case of a failure the listener 719 * may be called immediately, before TextToSpeech instance is fully constructed. 720 * @param engine Package name of the TTS engine to use. 721 */ TextToSpeech(Context context, OnInitListener listener, String engine)722 public TextToSpeech(Context context, OnInitListener listener, String engine) { 723 this(context, listener, engine, null, true); 724 } 725 726 /** 727 * Used by the framework to instantiate TextToSpeech objects with a supplied 728 * package name, instead of using {@link android.content.Context#getPackageName()} 729 * 730 * @hide 731 */ TextToSpeech(Context context, OnInitListener listener, String engine, String packageName, boolean useFallback)732 public TextToSpeech(Context context, OnInitListener listener, String engine, 733 String packageName, boolean useFallback) { 734 mContext = context; 735 mInitListener = listener; 736 mRequestedEngine = engine; 737 mUseFallback = useFallback; 738 739 mEarcons = new HashMap<String, Uri>(); 740 mUtterances = new HashMap<CharSequence, Uri>(); 741 mUtteranceProgressListener = null; 742 743 mEnginesHelper = new TtsEngines(mContext); 744 initTts(); 745 } 746 runActionNoReconnect(Action<R> action, R errorResult, String method, boolean onlyEstablishedConnection)747 private <R> R runActionNoReconnect(Action<R> action, R errorResult, String method, 748 boolean onlyEstablishedConnection) { 749 return runAction(action, errorResult, method, false, onlyEstablishedConnection); 750 } 751 runAction(Action<R> action, R errorResult, String method)752 private <R> R runAction(Action<R> action, R errorResult, String method) { 753 return runAction(action, errorResult, method, true, true); 754 } 755 runAction(Action<R> action, R errorResult, String method, boolean reconnect, boolean onlyEstablishedConnection)756 private <R> R runAction(Action<R> action, R errorResult, String method, 757 boolean reconnect, boolean onlyEstablishedConnection) { 758 synchronized (mStartLock) { 759 if (mServiceConnection == null) { 760 Log.w(TAG, method + " failed: not bound to TTS engine"); 761 return errorResult; 762 } 763 return mServiceConnection.runAction(action, errorResult, method, reconnect, 764 onlyEstablishedConnection); 765 } 766 } 767 initTts()768 private int initTts() { 769 // Step 1: Try connecting to the engine that was requested. 770 if (mRequestedEngine != null) { 771 if (mEnginesHelper.isEngineInstalled(mRequestedEngine)) { 772 if (connectToEngine(mRequestedEngine)) { 773 mCurrentEngine = mRequestedEngine; 774 return SUCCESS; 775 } else if (!mUseFallback) { 776 mCurrentEngine = null; 777 dispatchOnInit(ERROR); 778 return ERROR; 779 } 780 } else if (!mUseFallback) { 781 Log.i(TAG, "Requested engine not installed: " + mRequestedEngine); 782 mCurrentEngine = null; 783 dispatchOnInit(ERROR); 784 return ERROR; 785 } 786 } 787 788 // Step 2: Try connecting to the user's default engine. 789 final String defaultEngine = getDefaultEngine(); 790 if (defaultEngine != null && !defaultEngine.equals(mRequestedEngine)) { 791 if (connectToEngine(defaultEngine)) { 792 mCurrentEngine = defaultEngine; 793 return SUCCESS; 794 } 795 } 796 797 // Step 3: Try connecting to the highest ranked engine in the 798 // system. 799 final String highestRanked = mEnginesHelper.getHighestRankedEngineName(); 800 if (highestRanked != null && !highestRanked.equals(mRequestedEngine) && 801 !highestRanked.equals(defaultEngine)) { 802 if (connectToEngine(highestRanked)) { 803 mCurrentEngine = highestRanked; 804 return SUCCESS; 805 } 806 } 807 808 // NOTE: The API currently does not allow the caller to query whether 809 // they are actually connected to any engine. This might fail for various 810 // reasons like if the user disables all her TTS engines. 811 812 mCurrentEngine = null; 813 dispatchOnInit(ERROR); 814 return ERROR; 815 } 816 connectToEngine(String engine)817 private boolean connectToEngine(String engine) { 818 Connection connection = new Connection(); 819 Intent intent = new Intent(Engine.INTENT_ACTION_TTS_SERVICE); 820 intent.setPackage(engine); 821 boolean bound = mContext.bindService(intent, connection, Context.BIND_AUTO_CREATE); 822 if (!bound) { 823 Log.e(TAG, "Failed to bind to " + engine); 824 return false; 825 } else { 826 Log.i(TAG, "Sucessfully bound to " + engine); 827 mConnectingServiceConnection = connection; 828 return true; 829 } 830 } 831 dispatchOnInit(int result)832 private void dispatchOnInit(int result) { 833 synchronized (mStartLock) { 834 if (mInitListener != null) { 835 mInitListener.onInit(result); 836 mInitListener = null; 837 } 838 } 839 } 840 getCallerIdentity()841 private IBinder getCallerIdentity() { 842 return mServiceConnection.getCallerIdentity(); 843 } 844 845 /** 846 * Releases the resources used by the TextToSpeech engine. 847 * It is good practice for instance to call this method in the onDestroy() method of an Activity 848 * so the TextToSpeech engine can be cleanly stopped. 849 */ shutdown()850 public void shutdown() { 851 // Special case, we are asked to shutdown connection that did finalize its connection. 852 synchronized (mStartLock) { 853 if (mConnectingServiceConnection != null) { 854 mContext.unbindService(mConnectingServiceConnection); 855 mConnectingServiceConnection = null; 856 return; 857 } 858 } 859 860 // Post connection case 861 runActionNoReconnect(new Action<Void>() { 862 @Override 863 public Void run(ITextToSpeechService service) throws RemoteException { 864 service.setCallback(getCallerIdentity(), null); 865 service.stop(getCallerIdentity()); 866 mServiceConnection.disconnect(); 867 // Context#unbindService does not result in a call to 868 // ServiceConnection#onServiceDisconnected. As a result, the 869 // service ends up being destroyed (if there are no other open 870 // connections to it) but the process lives on and the 871 // ServiceConnection continues to refer to the destroyed service. 872 // 873 // This leads to tons of log spam about SynthThread being dead. 874 mServiceConnection = null; 875 mCurrentEngine = null; 876 return null; 877 } 878 }, null, "shutdown", false); 879 } 880 881 /** 882 * Adds a mapping between a string of text and a sound resource in a 883 * package. After a call to this method, subsequent calls to 884 * {@link #speak(CharSequence, int, Bundle, String)} will play the specified sound resource 885 * if it is available, or synthesize the text it is missing. 886 * 887 * @param text 888 * The string of text. Example: <code>"south_south_east"</code> 889 * 890 * @param packagename 891 * Pass the packagename of the application that contains the 892 * resource. If the resource is in your own application (this is 893 * the most common case), then put the packagename of your 894 * application here.<br/> 895 * Example: <b>"com.google.marvin.compass"</b><br/> 896 * The packagename can be found in the AndroidManifest.xml of 897 * your application. 898 * <p> 899 * <code><manifest xmlns:android="..." 900 * package="<b>com.google.marvin.compass</b>"></code> 901 * </p> 902 * 903 * @param resourceId 904 * Example: <code>R.raw.south_south_east</code> 905 * 906 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 907 */ addSpeech(String text, String packagename, @RawRes int resourceId)908 public int addSpeech(String text, String packagename, @RawRes int resourceId) { 909 synchronized (mStartLock) { 910 mUtterances.put(text, makeResourceUri(packagename, resourceId)); 911 return SUCCESS; 912 } 913 } 914 915 /** 916 * Adds a mapping between a CharSequence (may be spanned with TtsSpans) of text 917 * and a sound resource in a package. After a call to this method, subsequent calls to 918 * {@link #speak(CharSequence, int, Bundle, String)} will play the specified sound resource 919 * if it is available, or synthesize the text it is missing. 920 * 921 * @param text 922 * The string of text. Example: <code>"south_south_east"</code> 923 * 924 * @param packagename 925 * Pass the packagename of the application that contains the 926 * resource. If the resource is in your own application (this is 927 * the most common case), then put the packagename of your 928 * application here.<br/> 929 * Example: <b>"com.google.marvin.compass"</b><br/> 930 * The packagename can be found in the AndroidManifest.xml of 931 * your application. 932 * <p> 933 * <code><manifest xmlns:android="..." 934 * package="<b>com.google.marvin.compass</b>"></code> 935 * </p> 936 * 937 * @param resourceId 938 * Example: <code>R.raw.south_south_east</code> 939 * 940 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 941 */ addSpeech(CharSequence text, String packagename, @RawRes int resourceId)942 public int addSpeech(CharSequence text, String packagename, @RawRes int resourceId) { 943 synchronized (mStartLock) { 944 mUtterances.put(text, makeResourceUri(packagename, resourceId)); 945 return SUCCESS; 946 } 947 } 948 949 /** 950 * Adds a mapping between a string of text and a sound file. Using this, it is possible to 951 * add custom pronounciations for a string of text. After a call to this method, subsequent 952 * calls to {@link #speak(CharSequence, int, Bundle, String)} will play the specified sound 953 * resource if it is available, or synthesize the text it is missing. 954 * 955 * @param text 956 * The string of text. Example: <code>"south_south_east"</code> 957 * @param filename 958 * The full path to the sound file (for example: 959 * "/sdcard/mysounds/hello.wav") 960 * 961 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 962 */ addSpeech(String text, String filename)963 public int addSpeech(String text, String filename) { 964 synchronized (mStartLock) { 965 mUtterances.put(text, Uri.parse(filename)); 966 return SUCCESS; 967 } 968 } 969 970 /** 971 * Adds a mapping between a CharSequence (may be spanned with TtsSpans and a sound file. 972 * Using this, it is possible to add custom pronounciations for a string of text. After a call 973 * to this method, subsequent calls to {@link #speak(CharSequence, int, Bundle, String)} 974 * will play the specified sound resource if it is available, or synthesize the text it is 975 * missing. 976 * 977 * @param text 978 * The string of text. Example: <code>"south_south_east"</code> 979 * @param file 980 * File object pointing to the sound file. 981 * 982 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 983 */ addSpeech(CharSequence text, File file)984 public int addSpeech(CharSequence text, File file) { 985 synchronized (mStartLock) { 986 mUtterances.put(text, Uri.fromFile(file)); 987 return SUCCESS; 988 } 989 } 990 991 /** 992 * Adds a mapping between a string of text and a sound resource in a 993 * package. Use this to add custom earcons. 994 * 995 * @see #playEarcon(String, int, HashMap) 996 * 997 * @param earcon The name of the earcon. 998 * Example: <code>"[tick]"</code><br/> 999 * 1000 * @param packagename 1001 * the package name of the application that contains the 1002 * resource. This can for instance be the package name of your own application. 1003 * Example: <b>"com.google.marvin.compass"</b><br/> 1004 * The package name can be found in the AndroidManifest.xml of 1005 * the application containing the resource. 1006 * <p> 1007 * <code><manifest xmlns:android="..." 1008 * package="<b>com.google.marvin.compass</b>"></code> 1009 * </p> 1010 * 1011 * @param resourceId 1012 * Example: <code>R.raw.tick_snd</code> 1013 * 1014 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 1015 */ addEarcon(String earcon, String packagename, @RawRes int resourceId)1016 public int addEarcon(String earcon, String packagename, @RawRes int resourceId) { 1017 synchronized(mStartLock) { 1018 mEarcons.put(earcon, makeResourceUri(packagename, resourceId)); 1019 return SUCCESS; 1020 } 1021 } 1022 1023 /** 1024 * Adds a mapping between a string of text and a sound file. 1025 * Use this to add custom earcons. 1026 * 1027 * @see #playEarcon(String, int, HashMap) 1028 * 1029 * @param earcon 1030 * The name of the earcon. 1031 * Example: <code>"[tick]"</code> 1032 * @param filename 1033 * The full path to the sound file (for example: 1034 * "/sdcard/mysounds/tick.wav") 1035 * 1036 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 1037 * 1038 * @deprecated As of API level 21, replaced by 1039 * {@link #addEarcon(String, File)}. 1040 */ 1041 @Deprecated addEarcon(String earcon, String filename)1042 public int addEarcon(String earcon, String filename) { 1043 synchronized(mStartLock) { 1044 mEarcons.put(earcon, Uri.parse(filename)); 1045 return SUCCESS; 1046 } 1047 } 1048 1049 /** 1050 * Adds a mapping between a string of text and a sound file. 1051 * Use this to add custom earcons. 1052 * 1053 * @see #playEarcon(String, int, HashMap) 1054 * 1055 * @param earcon 1056 * The name of the earcon. 1057 * Example: <code>"[tick]"</code> 1058 * @param file 1059 * File object pointing to the sound file. 1060 * 1061 * @return Code indicating success or failure. See {@link #ERROR} and {@link #SUCCESS}. 1062 */ addEarcon(String earcon, File file)1063 public int addEarcon(String earcon, File file) { 1064 synchronized(mStartLock) { 1065 mEarcons.put(earcon, Uri.fromFile(file)); 1066 return SUCCESS; 1067 } 1068 } 1069 makeResourceUri(String packageName, int resourceId)1070 private Uri makeResourceUri(String packageName, int resourceId) { 1071 return new Uri.Builder() 1072 .scheme(ContentResolver.SCHEME_ANDROID_RESOURCE) 1073 .encodedAuthority(packageName) 1074 .appendEncodedPath(String.valueOf(resourceId)) 1075 .build(); 1076 } 1077 1078 /** 1079 * Speaks the text using the specified queuing strategy and speech parameters, the text may 1080 * be spanned with TtsSpans. 1081 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1082 * requests and then returns. The synthesis might not have finished (or even started!) at the 1083 * time when this method returns. In order to reliably detect errors during synthesis, 1084 * we recommend setting an utterance progress listener (see 1085 * {@link #setOnUtteranceProgressListener}) and using the 1086 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1087 * 1088 * @param text The string of text to be spoken. No longer than 1089 * {@link #getMaxSpeechInputLength()} characters. 1090 * @param queueMode The queuing strategy to use, {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1091 * @param params Parameters for the request. Can be null. 1092 * Supported parameter names: 1093 * {@link Engine#KEY_PARAM_STREAM}, 1094 * {@link Engine#KEY_PARAM_VOLUME}, 1095 * {@link Engine#KEY_PARAM_PAN}. 1096 * Engine specific parameters may be passed in but the parameter keys 1097 * must be prefixed by the name of the engine they are intended for. For example 1098 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1099 * engine named "com.svox.pico" if it is being used. 1100 * @param utteranceId An unique identifier for this request. 1101 * 1102 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the speak operation. 1103 */ speak(final CharSequence text, final int queueMode, final Bundle params, final String utteranceId)1104 public int speak(final CharSequence text, 1105 final int queueMode, 1106 final Bundle params, 1107 final String utteranceId) { 1108 return runAction(new Action<Integer>() { 1109 @Override 1110 public Integer run(ITextToSpeechService service) throws RemoteException { 1111 Uri utteranceUri = mUtterances.get(text); 1112 if (utteranceUri != null) { 1113 return service.playAudio(getCallerIdentity(), utteranceUri, queueMode, 1114 getParams(params), utteranceId); 1115 } else { 1116 return service.speak(getCallerIdentity(), text, queueMode, getParams(params), 1117 utteranceId); 1118 } 1119 } 1120 }, ERROR, "speak"); 1121 } 1122 1123 /** 1124 * Speaks the string using the specified queuing strategy and speech parameters. 1125 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1126 * requests and then returns. The synthesis might not have finished (or even started!) at the 1127 * time when this method returns. In order to reliably detect errors during synthesis, 1128 * we recommend setting an utterance progress listener (see 1129 * {@link #setOnUtteranceProgressListener}) and using the 1130 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1131 * 1132 * @param text The string of text to be spoken. No longer than 1133 * {@link #getMaxSpeechInputLength()} characters. 1134 * @param queueMode The queuing strategy to use, {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1135 * @param params Parameters for the request. Can be null. 1136 * Supported parameter names: 1137 * {@link Engine#KEY_PARAM_STREAM}, 1138 * {@link Engine#KEY_PARAM_UTTERANCE_ID}, 1139 * {@link Engine#KEY_PARAM_VOLUME}, 1140 * {@link Engine#KEY_PARAM_PAN}. 1141 * Engine specific parameters may be passed in but the parameter keys 1142 * must be prefixed by the name of the engine they are intended for. For example 1143 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1144 * engine named "com.svox.pico" if it is being used. 1145 * 1146 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the speak operation. 1147 * @deprecated As of API level 21, replaced by 1148 * {@link #speak(CharSequence, int, Bundle, String)}. 1149 */ 1150 @Deprecated 1151 public int speak(final String text, final int queueMode, final HashMap<String, String> params) { 1152 return speak(text, queueMode, convertParamsHashMaptoBundle(params), 1153 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1154 } 1155 1156 /** 1157 * Plays the earcon using the specified queueing mode and parameters. 1158 * The earcon must already have been added with {@link #addEarcon(String, String)} or 1159 * {@link #addEarcon(String, String, int)}. 1160 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1161 * requests and then returns. The synthesis might not have finished (or even started!) at the 1162 * time when this method returns. In order to reliably detect errors during synthesis, 1163 * we recommend setting an utterance progress listener (see 1164 * {@link #setOnUtteranceProgressListener}) and using the 1165 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1166 * 1167 * @param earcon The earcon that should be played 1168 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1169 * @param params Parameters for the request. Can be null. 1170 * Supported parameter names: 1171 * {@link Engine#KEY_PARAM_STREAM}, 1172 * Engine specific parameters may be passed in but the parameter keys 1173 * must be prefixed by the name of the engine they are intended for. For example 1174 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1175 * engine named "com.svox.pico" if it is being used. 1176 * 1177 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playEarcon operation. 1178 */ 1179 public int playEarcon(final String earcon, final int queueMode, 1180 final Bundle params, final String utteranceId) { 1181 return runAction(new Action<Integer>() { 1182 @Override 1183 public Integer run(ITextToSpeechService service) throws RemoteException { 1184 Uri earconUri = mEarcons.get(earcon); 1185 if (earconUri == null) { 1186 return ERROR; 1187 } 1188 return service.playAudio(getCallerIdentity(), earconUri, queueMode, 1189 getParams(params), utteranceId); 1190 } 1191 }, ERROR, "playEarcon"); 1192 } 1193 1194 /** 1195 * Plays the earcon using the specified queueing mode and parameters. 1196 * The earcon must already have been added with {@link #addEarcon(String, String)} or 1197 * {@link #addEarcon(String, String, int)}. 1198 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1199 * requests and then returns. The synthesis might not have finished (or even started!) at the 1200 * time when this method returns. In order to reliably detect errors during synthesis, 1201 * we recommend setting an utterance progress listener (see 1202 * {@link #setOnUtteranceProgressListener}) and using the 1203 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1204 * 1205 * @param earcon The earcon that should be played 1206 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1207 * @param params Parameters for the request. Can be null. 1208 * Supported parameter names: 1209 * {@link Engine#KEY_PARAM_STREAM}, 1210 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1211 * Engine specific parameters may be passed in but the parameter keys 1212 * must be prefixed by the name of the engine they are intended for. For example 1213 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1214 * engine named "com.svox.pico" if it is being used. 1215 * 1216 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playEarcon operation. 1217 * @deprecated As of API level 21, replaced by 1218 * {@link #playEarcon(String, int, Bundle, String)}. 1219 */ 1220 @Deprecated 1221 public int playEarcon(final String earcon, final int queueMode, 1222 final HashMap<String, String> params) { 1223 return playEarcon(earcon, queueMode, convertParamsHashMaptoBundle(params), 1224 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1225 } 1226 1227 /** 1228 * Plays silence for the specified amount of time using the specified 1229 * queue mode. 1230 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1231 * requests and then returns. The synthesis might not have finished (or even started!) at the 1232 * time when this method returns. In order to reliably detect errors during synthesis, 1233 * we recommend setting an utterance progress listener (see 1234 * {@link #setOnUtteranceProgressListener}) and using the 1235 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1236 * 1237 * @param durationInMs The duration of the silence. 1238 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1239 * @param utteranceId An unique identifier for this request. 1240 * 1241 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playSilentUtterance operation. 1242 */ 1243 public int playSilentUtterance(final long durationInMs, final int queueMode, 1244 final String utteranceId) { 1245 return runAction(new Action<Integer>() { 1246 @Override 1247 public Integer run(ITextToSpeechService service) throws RemoteException { 1248 return service.playSilence(getCallerIdentity(), durationInMs, 1249 queueMode, utteranceId); 1250 } 1251 }, ERROR, "playSilentUtterance"); 1252 } 1253 1254 /** 1255 * Plays silence for the specified amount of time using the specified 1256 * queue mode. 1257 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1258 * requests and then returns. The synthesis might not have finished (or even started!) at the 1259 * time when this method returns. In order to reliably detect errors during synthesis, 1260 * we recommend setting an utterance progress listener (see 1261 * {@link #setOnUtteranceProgressListener}) and using the 1262 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1263 * 1264 * @param durationInMs The duration of the silence. 1265 * @param queueMode {@link #QUEUE_ADD} or {@link #QUEUE_FLUSH}. 1266 * @param params Parameters for the request. Can be null. 1267 * Supported parameter names: 1268 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1269 * Engine specific parameters may be passed in but the parameter keys 1270 * must be prefixed by the name of the engine they are intended for. For example 1271 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1272 * engine named "com.svox.pico" if it is being used. 1273 * 1274 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the playSilence operation. 1275 * @deprecated As of API level 21, replaced by 1276 * {@link #playSilentUtterance(long, int, String)}. 1277 */ 1278 @Deprecated 1279 public int playSilence(final long durationInMs, final int queueMode, 1280 final HashMap<String, String> params) { 1281 return playSilentUtterance(durationInMs, queueMode, 1282 params == null ? null : params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1283 } 1284 1285 /** 1286 * Queries the engine for the set of features it supports for a given locale. 1287 * Features can either be framework defined, e.g. 1288 * {@link TextToSpeech.Engine#KEY_FEATURE_NETWORK_SYNTHESIS} or engine specific. 1289 * Engine specific keys must be prefixed by the name of the engine they 1290 * are intended for. These keys can be used as parameters to 1291 * {@link TextToSpeech#speak(String, int, java.util.HashMap)} and 1292 * {@link TextToSpeech#synthesizeToFile(String, java.util.HashMap, String)}. 1293 * 1294 * Features values are strings and their values must meet restrictions described in their 1295 * documentation. 1296 * 1297 * @param locale The locale to query features for. 1298 * @return Set instance. May return {@code null} on error. 1299 * @deprecated As of API level 21, please use voices. In order to query features of the voice, 1300 * call {@link #getVoices()} to retrieve the list of available voices and 1301 * {@link Voice#getFeatures()} to retrieve the set of features. 1302 */ 1303 @Deprecated 1304 public Set<String> getFeatures(final Locale locale) { 1305 return runAction(new Action<Set<String>>() { 1306 @Override 1307 public Set<String> run(ITextToSpeechService service) throws RemoteException { 1308 String[] features = null; 1309 try { 1310 features = service.getFeaturesForLanguage( 1311 locale.getISO3Language(), locale.getISO3Country(), locale.getVariant()); 1312 } catch(MissingResourceException e) { 1313 Log.w(TAG, "Couldn't retrieve 3 letter ISO 639-2/T language and/or ISO 3166 " + 1314 "country code for locale: " + locale, e); 1315 return null; 1316 } 1317 1318 if (features != null) { 1319 final Set<String> featureSet = new HashSet<String>(); 1320 Collections.addAll(featureSet, features); 1321 return featureSet; 1322 } 1323 return null; 1324 } 1325 }, null, "getFeatures"); 1326 } 1327 1328 /** 1329 * Checks whether the TTS engine is busy speaking. Note that a speech item is 1330 * considered complete once it's audio data has been sent to the audio mixer, or 1331 * written to a file. There might be a finite lag between this point, and when 1332 * the audio hardware completes playback. 1333 * 1334 * @return {@code true} if the TTS engine is speaking. 1335 */ 1336 public boolean isSpeaking() { 1337 return runAction(new Action<Boolean>() { 1338 @Override 1339 public Boolean run(ITextToSpeechService service) throws RemoteException { 1340 return service.isSpeaking(); 1341 } 1342 }, false, "isSpeaking"); 1343 } 1344 1345 /** 1346 * Interrupts the current utterance (whether played or rendered to file) and discards other 1347 * utterances in the queue. 1348 * 1349 * @return {@link #ERROR} or {@link #SUCCESS}. 1350 */ 1351 public int stop() { 1352 return runAction(new Action<Integer>() { 1353 @Override 1354 public Integer run(ITextToSpeechService service) throws RemoteException { 1355 return service.stop(getCallerIdentity()); 1356 } 1357 }, ERROR, "stop"); 1358 } 1359 1360 /** 1361 * Sets the speech rate. 1362 * 1363 * This has no effect on any pre-recorded speech. 1364 * 1365 * @param speechRate Speech rate. {@code 1.0} is the normal speech rate, 1366 * lower values slow down the speech ({@code 0.5} is half the normal speech rate), 1367 * greater values accelerate it ({@code 2.0} is twice the normal speech rate). 1368 * 1369 * @return {@link #ERROR} or {@link #SUCCESS}. 1370 */ 1371 public int setSpeechRate(float speechRate) { 1372 if (speechRate > 0.0f) { 1373 int intRate = (int)(speechRate * 100); 1374 if (intRate > 0) { 1375 synchronized (mStartLock) { 1376 mParams.putInt(Engine.KEY_PARAM_RATE, intRate); 1377 } 1378 return SUCCESS; 1379 } 1380 } 1381 return ERROR; 1382 } 1383 1384 /** 1385 * Sets the speech pitch for the TextToSpeech engine. 1386 * 1387 * This has no effect on any pre-recorded speech. 1388 * 1389 * @param pitch Speech pitch. {@code 1.0} is the normal pitch, 1390 * lower values lower the tone of the synthesized voice, 1391 * greater values increase it. 1392 * 1393 * @return {@link #ERROR} or {@link #SUCCESS}. 1394 */ 1395 public int setPitch(float pitch) { 1396 if (pitch > 0.0f) { 1397 int intPitch = (int)(pitch * 100); 1398 if (intPitch > 0) { 1399 synchronized (mStartLock) { 1400 mParams.putInt(Engine.KEY_PARAM_PITCH, intPitch); 1401 } 1402 return SUCCESS; 1403 } 1404 } 1405 return ERROR; 1406 } 1407 1408 /** 1409 * Sets the audio attributes to be used when speaking text or playing 1410 * back a file. 1411 * 1412 * @param audioAttributes Valid AudioAttributes instance. 1413 * 1414 * @return {@link #ERROR} or {@link #SUCCESS}. 1415 */ 1416 public int setAudioAttributes(AudioAttributes audioAttributes) { 1417 if (audioAttributes != null) { 1418 synchronized (mStartLock) { 1419 mParams.putParcelable(Engine.KEY_PARAM_AUDIO_ATTRIBUTES, 1420 audioAttributes); 1421 } 1422 return SUCCESS; 1423 } 1424 return ERROR; 1425 } 1426 1427 /** 1428 * @return the engine currently in use by this TextToSpeech instance. 1429 * @hide 1430 */ 1431 @UnsupportedAppUsage 1432 public String getCurrentEngine() { 1433 return mCurrentEngine; 1434 } 1435 1436 /** 1437 * Returns a Locale instance describing the language currently being used as the default 1438 * Text-to-speech language. 1439 * 1440 * The locale object returned by this method is NOT a valid one. It has identical form to the 1441 * one in {@link #getLanguage()}. Please refer to {@link #getLanguage()} for more information. 1442 * 1443 * @return language, country (if any) and variant (if any) used by the client stored in a 1444 * Locale instance, or {@code null} on error. 1445 * @deprecated As of API level 21, use <code>getDefaultVoice().getLocale()</code> ({@link 1446 * #getDefaultVoice()}) 1447 */ 1448 @Deprecated 1449 public Locale getDefaultLanguage() { 1450 return runAction(new Action<Locale>() { 1451 @Override 1452 public Locale run(ITextToSpeechService service) throws RemoteException { 1453 String[] defaultLanguage = service.getClientDefaultLanguage(); 1454 1455 return new Locale(defaultLanguage[0], defaultLanguage[1], defaultLanguage[2]); 1456 } 1457 }, null, "getDefaultLanguage"); 1458 } 1459 1460 /** 1461 * Sets the text-to-speech language. 1462 * The TTS engine will try to use the closest match to the specified 1463 * language as represented by the Locale, but there is no guarantee that the exact same Locale 1464 * will be used. Use {@link #isLanguageAvailable(Locale)} to check the level of support 1465 * before choosing the language to use for the next utterances. 1466 * 1467 * This method sets the current voice to the default one for the given Locale; 1468 * {@link #getVoice()} can be used to retrieve it. 1469 * 1470 * @param loc The locale describing the language to be used. 1471 * 1472 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1473 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1474 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1475 */ 1476 public int setLanguage(final Locale loc) { 1477 return runAction(new Action<Integer>() { 1478 @Override 1479 public Integer run(ITextToSpeechService service) throws RemoteException { 1480 if (loc == null) { 1481 return LANG_NOT_SUPPORTED; 1482 } 1483 String language = null, country = null; 1484 try { 1485 language = loc.getISO3Language(); 1486 } catch (MissingResourceException e) { 1487 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1488 return LANG_NOT_SUPPORTED; 1489 } 1490 1491 try { 1492 country = loc.getISO3Country(); 1493 } catch (MissingResourceException e) { 1494 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1495 return LANG_NOT_SUPPORTED; 1496 } 1497 1498 String variant = loc.getVariant(); 1499 1500 // As of API level 21, setLanguage is implemented using setVoice. 1501 // (which, in the default implementation, will call loadLanguage on the service 1502 // interface). 1503 1504 // Sanitize locale using isLanguageAvailable. 1505 int result = service.isLanguageAvailable(language, country, variant); 1506 if (result >= LANG_AVAILABLE) { 1507 // Get the default voice for the locale. 1508 String voiceName = service.getDefaultVoiceNameFor(language, country, variant); 1509 if (TextUtils.isEmpty(voiceName)) { 1510 Log.w(TAG, "Couldn't find the default voice for " + language + "-" + 1511 country + "-" + variant); 1512 return LANG_NOT_SUPPORTED; 1513 } 1514 1515 // Load it. 1516 if (service.loadVoice(getCallerIdentity(), voiceName) == TextToSpeech.ERROR) { 1517 Log.w(TAG, "The service claimed " + language + "-" + country + "-" 1518 + variant + " was available with voice name " + voiceName 1519 + " but loadVoice returned ERROR"); 1520 return LANG_NOT_SUPPORTED; 1521 } 1522 1523 // Set the language/country/variant of the voice, so #getLanguage will return 1524 // the currently set voice locale when called. 1525 Voice voice = getVoice(service, voiceName); 1526 if (voice == null) { 1527 Log.w(TAG, "getDefaultVoiceNameFor returned " + voiceName + " for locale " 1528 + language + "-" + country + "-" + variant 1529 + " but getVoice returns null"); 1530 return LANG_NOT_SUPPORTED; 1531 } 1532 String voiceLanguage = ""; 1533 try { 1534 voiceLanguage = voice.getLocale().getISO3Language(); 1535 } catch (MissingResourceException e) { 1536 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + 1537 voice.getLocale(), e); 1538 } 1539 1540 String voiceCountry = ""; 1541 try { 1542 voiceCountry = voice.getLocale().getISO3Country(); 1543 } catch (MissingResourceException e) { 1544 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + 1545 voice.getLocale(), e); 1546 } 1547 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, voiceName); 1548 mParams.putString(Engine.KEY_PARAM_LANGUAGE, voiceLanguage); 1549 mParams.putString(Engine.KEY_PARAM_COUNTRY, voiceCountry); 1550 mParams.putString(Engine.KEY_PARAM_VARIANT, voice.getLocale().getVariant()); 1551 } 1552 return result; 1553 } 1554 }, LANG_NOT_SUPPORTED, "setLanguage"); 1555 } 1556 1557 /** 1558 * Returns a Locale instance describing the language currently being used for synthesis 1559 * requests sent to the TextToSpeech engine. 1560 * 1561 * In Android 4.2 and before (API <= 17) this function returns the language that is currently 1562 * being used by the TTS engine. That is the last language set by this or any other 1563 * client by a {@link TextToSpeech#setLanguage} call to the same engine. 1564 * 1565 * In Android versions after 4.2 this function returns the language that is currently being 1566 * used for the synthesis requests sent from this client. That is the last language set 1567 * by a {@link TextToSpeech#setLanguage} call on this instance. 1568 * 1569 * If a voice is set (by {@link #setVoice(Voice)}), getLanguage will return the language of 1570 * the currently set voice. 1571 * 1572 * Please note that the Locale object returned by this method is NOT a valid Locale object. Its 1573 * language field contains a three-letter ISO 639-2/T code (where a proper Locale would use 1574 * a two-letter ISO 639-1 code), and the country field contains a three-letter ISO 3166 country 1575 * code (where a proper Locale would use a two-letter ISO 3166-1 code). 1576 * 1577 * @return language, country (if any) and variant (if any) used by the client stored in a 1578 * Locale instance, or {@code null} on error. 1579 * 1580 * @deprecated As of API level 21, please use <code>getVoice().getLocale()</code> 1581 * ({@link #getVoice()}). 1582 */ 1583 @Deprecated 1584 public Locale getLanguage() { 1585 return runAction(new Action<Locale>() { 1586 @Override 1587 public Locale run(ITextToSpeechService service) { 1588 /* No service call, but we're accessing mParams, hence need for 1589 wrapping it as an Action instance */ 1590 String lang = mParams.getString(Engine.KEY_PARAM_LANGUAGE, ""); 1591 String country = mParams.getString(Engine.KEY_PARAM_COUNTRY, ""); 1592 String variant = mParams.getString(Engine.KEY_PARAM_VARIANT, ""); 1593 return new Locale(lang, country, variant); 1594 } 1595 }, null, "getLanguage"); 1596 } 1597 1598 /** 1599 * Query the engine about the set of available languages. 1600 */ 1601 public Set<Locale> getAvailableLanguages() { 1602 return runAction(new Action<Set<Locale>>() { 1603 @Override 1604 public Set<Locale> run(ITextToSpeechService service) throws RemoteException { 1605 List<Voice> voices = service.getVoices(); 1606 if (voices == null) { 1607 return new HashSet<Locale>(); 1608 } 1609 HashSet<Locale> locales = new HashSet<Locale>(); 1610 for (Voice voice : voices) { 1611 locales.add(voice.getLocale()); 1612 } 1613 return locales; 1614 } 1615 }, null, "getAvailableLanguages"); 1616 } 1617 1618 /** 1619 * Query the engine about the set of available voices. 1620 * 1621 * Each TTS Engine can expose multiple voices for each locale, each with a different set of 1622 * features. 1623 * 1624 * @see #setVoice(Voice) 1625 * @see Voice 1626 */ 1627 public Set<Voice> getVoices() { 1628 return runAction(new Action<Set<Voice>>() { 1629 @Override 1630 public Set<Voice> run(ITextToSpeechService service) throws RemoteException { 1631 List<Voice> voices = service.getVoices(); 1632 return (voices != null) ? new HashSet<Voice>(voices) : new HashSet<Voice>(); 1633 } 1634 }, null, "getVoices"); 1635 } 1636 1637 /** 1638 * Sets the text-to-speech voice. 1639 * 1640 * @param voice One of objects returned by {@link #getVoices()}. 1641 * 1642 * @return {@link #ERROR} or {@link #SUCCESS}. 1643 * 1644 * @see #getVoices 1645 * @see Voice 1646 */ 1647 public int setVoice(final Voice voice) { 1648 return runAction(new Action<Integer>() { 1649 @Override 1650 public Integer run(ITextToSpeechService service) throws RemoteException { 1651 int result = service.loadVoice(getCallerIdentity(), voice.getName()); 1652 if (result == SUCCESS) { 1653 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, voice.getName()); 1654 1655 // Set the language/country/variant, so #getLanguage will return the voice 1656 // locale when called. 1657 String language = ""; 1658 try { 1659 language = voice.getLocale().getISO3Language(); 1660 } catch (MissingResourceException e) { 1661 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + 1662 voice.getLocale(), e); 1663 } 1664 1665 String country = ""; 1666 try { 1667 country = voice.getLocale().getISO3Country(); 1668 } catch (MissingResourceException e) { 1669 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + 1670 voice.getLocale(), e); 1671 } 1672 mParams.putString(Engine.KEY_PARAM_LANGUAGE, language); 1673 mParams.putString(Engine.KEY_PARAM_COUNTRY, country); 1674 mParams.putString(Engine.KEY_PARAM_VARIANT, voice.getLocale().getVariant()); 1675 } 1676 return result; 1677 } 1678 }, LANG_NOT_SUPPORTED, "setVoice"); 1679 } 1680 1681 /** 1682 * Returns a Voice instance describing the voice currently being used for synthesis 1683 * requests sent to the TextToSpeech engine. 1684 * 1685 * @return Voice instance used by the client, or {@code null} if not set or on error. 1686 * 1687 * @see #getVoices 1688 * @see #setVoice 1689 * @see Voice 1690 */ 1691 public Voice getVoice() { 1692 return runAction(new Action<Voice>() { 1693 @Override 1694 public Voice run(ITextToSpeechService service) throws RemoteException { 1695 String voiceName = mParams.getString(Engine.KEY_PARAM_VOICE_NAME, ""); 1696 if (TextUtils.isEmpty(voiceName)) { 1697 return null; 1698 } 1699 return getVoice(service, voiceName); 1700 } 1701 }, null, "getVoice"); 1702 } 1703 1704 1705 /** 1706 * Returns a Voice instance of the voice with the given voice name. 1707 * 1708 * @return Voice instance with the given voice name, or {@code null} if not set or on error. 1709 * 1710 * @see Voice 1711 */ 1712 private Voice getVoice(ITextToSpeechService service, String voiceName) throws RemoteException { 1713 List<Voice> voices = service.getVoices(); 1714 if (voices == null) { 1715 Log.w(TAG, "getVoices returned null"); 1716 return null; 1717 } 1718 for (Voice voice : voices) { 1719 if (voice.getName().equals(voiceName)) { 1720 return voice; 1721 } 1722 } 1723 Log.w(TAG, "Could not find voice " + voiceName + " in voice list"); 1724 return null; 1725 } 1726 1727 /** 1728 * Returns a Voice instance that's the default voice for the default Text-to-speech language. 1729 * @return The default voice instance for the default language, or {@code null} if not set or 1730 * on error. 1731 */ 1732 public Voice getDefaultVoice() { 1733 return runAction(new Action<Voice>() { 1734 @Override 1735 public Voice run(ITextToSpeechService service) throws RemoteException { 1736 1737 String[] defaultLanguage = service.getClientDefaultLanguage(); 1738 1739 if (defaultLanguage == null || defaultLanguage.length == 0) { 1740 Log.e(TAG, "service.getClientDefaultLanguage() returned empty array"); 1741 return null; 1742 } 1743 String language = defaultLanguage[0]; 1744 String country = (defaultLanguage.length > 1) ? defaultLanguage[1] : ""; 1745 String variant = (defaultLanguage.length > 2) ? defaultLanguage[2] : ""; 1746 1747 // Sanitize the locale using isLanguageAvailable. 1748 int result = service.isLanguageAvailable(language, country, variant); 1749 if (result < LANG_AVAILABLE) { 1750 // The default language is not supported. 1751 return null; 1752 } 1753 1754 // Get the default voice name 1755 String voiceName = service.getDefaultVoiceNameFor(language, country, variant); 1756 if (TextUtils.isEmpty(voiceName)) { 1757 return null; 1758 } 1759 1760 // Find it 1761 List<Voice> voices = service.getVoices(); 1762 if (voices == null) { 1763 return null; 1764 } 1765 for (Voice voice : voices) { 1766 if (voice.getName().equals(voiceName)) { 1767 return voice; 1768 } 1769 } 1770 return null; 1771 } 1772 }, null, "getDefaultVoice"); 1773 } 1774 1775 1776 1777 /** 1778 * Checks if the specified language as represented by the Locale is available and supported. 1779 * 1780 * @param loc The Locale describing the language to be used. 1781 * 1782 * @return Code indicating the support status for the locale. See {@link #LANG_AVAILABLE}, 1783 * {@link #LANG_COUNTRY_AVAILABLE}, {@link #LANG_COUNTRY_VAR_AVAILABLE}, 1784 * {@link #LANG_MISSING_DATA} and {@link #LANG_NOT_SUPPORTED}. 1785 */ 1786 public int isLanguageAvailable(final Locale loc) { 1787 return runAction(new Action<Integer>() { 1788 @Override 1789 public Integer run(ITextToSpeechService service) throws RemoteException { 1790 String language = null, country = null; 1791 1792 try { 1793 language = loc.getISO3Language(); 1794 } catch (MissingResourceException e) { 1795 Log.w(TAG, "Couldn't retrieve ISO 639-2/T language code for locale: " + loc, e); 1796 return LANG_NOT_SUPPORTED; 1797 } 1798 1799 try { 1800 country = loc.getISO3Country(); 1801 } catch (MissingResourceException e) { 1802 Log.w(TAG, "Couldn't retrieve ISO 3166 country code for locale: " + loc, e); 1803 return LANG_NOT_SUPPORTED; 1804 } 1805 1806 return service.isLanguageAvailable(language, country, loc.getVariant()); 1807 } 1808 }, LANG_NOT_SUPPORTED, "isLanguageAvailable"); 1809 } 1810 1811 /** 1812 * Synthesizes the given text to a file using the specified parameters. 1813 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1814 * requests and then returns. The synthesis might not have finished (or even started!) at the 1815 * time when this method returns. In order to reliably detect errors during synthesis, 1816 * we recommend setting an utterance progress listener (see 1817 * {@link #setOnUtteranceProgressListener}). 1818 * 1819 * @param text The text that should be synthesized. No longer than 1820 * {@link #getMaxSpeechInputLength()} characters. 1821 * @param params Parameters for the request. Can be null. 1822 * Engine specific parameters may be passed in but the parameter keys 1823 * must be prefixed by the name of the engine they are intended for. For example 1824 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1825 * engine named "com.svox.pico" if it is being used. 1826 * @param file File to write the generated audio data to. 1827 * @param utteranceId An unique identifier for this request. 1828 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the synthesizeToFile operation. 1829 */ 1830 public int synthesizeToFile(final CharSequence text, final Bundle params, 1831 final File file, final String utteranceId) { 1832 return runAction(new Action<Integer>() { 1833 @Override 1834 public Integer run(ITextToSpeechService service) throws RemoteException { 1835 ParcelFileDescriptor fileDescriptor; 1836 int returnValue; 1837 try { 1838 if(file.exists() && !file.canWrite()) { 1839 Log.e(TAG, "Can't write to " + file); 1840 return ERROR; 1841 } 1842 fileDescriptor = ParcelFileDescriptor.open(file, 1843 ParcelFileDescriptor.MODE_WRITE_ONLY | 1844 ParcelFileDescriptor.MODE_CREATE | 1845 ParcelFileDescriptor.MODE_TRUNCATE); 1846 returnValue = service.synthesizeToFileDescriptor(getCallerIdentity(), text, 1847 fileDescriptor, getParams(params), utteranceId); 1848 fileDescriptor.close(); 1849 return returnValue; 1850 } catch (FileNotFoundException e) { 1851 Log.e(TAG, "Opening file " + file + " failed", e); 1852 return ERROR; 1853 } catch (IOException e) { 1854 Log.e(TAG, "Closing file " + file + " failed", e); 1855 return ERROR; 1856 } 1857 } 1858 }, ERROR, "synthesizeToFile"); 1859 } 1860 1861 /** 1862 * Synthesizes the given text to a file using the specified parameters. 1863 * This method is asynchronous, i.e. the method just adds the request to the queue of TTS 1864 * requests and then returns. The synthesis might not have finished (or even started!) at the 1865 * time when this method returns. In order to reliably detect errors during synthesis, 1866 * we recommend setting an utterance progress listener (see 1867 * {@link #setOnUtteranceProgressListener}) and using the 1868 * {@link Engine#KEY_PARAM_UTTERANCE_ID} parameter. 1869 * 1870 * @param text The text that should be synthesized. No longer than 1871 * {@link #getMaxSpeechInputLength()} characters. 1872 * @param params Parameters for the request. Can be null. 1873 * Supported parameter names: 1874 * {@link Engine#KEY_PARAM_UTTERANCE_ID}. 1875 * Engine specific parameters may be passed in but the parameter keys 1876 * must be prefixed by the name of the engine they are intended for. For example 1877 * the keys "com.svox.pico_foo" and "com.svox.pico:bar" will be passed to the 1878 * engine named "com.svox.pico" if it is being used. 1879 * @param filename Absolute file filename to write the generated audio data to.It should be 1880 * something like "/sdcard/myappsounds/mysound.wav". 1881 * 1882 * @return {@link #ERROR} or {@link #SUCCESS} of <b>queuing</b> the synthesizeToFile operation. 1883 * @deprecated As of API level 21, replaced by 1884 * {@link #synthesizeToFile(CharSequence, Bundle, File, String)}. 1885 */ 1886 @Deprecated 1887 public int synthesizeToFile(final String text, final HashMap<String, String> params, 1888 final String filename) { 1889 return synthesizeToFile(text, convertParamsHashMaptoBundle(params), 1890 new File(filename), params.get(Engine.KEY_PARAM_UTTERANCE_ID)); 1891 } 1892 1893 private Bundle convertParamsHashMaptoBundle(HashMap<String, String> params) { 1894 if (params != null && !params.isEmpty()) { 1895 Bundle bundle = new Bundle(); 1896 copyIntParam(bundle, params, Engine.KEY_PARAM_STREAM); 1897 copyIntParam(bundle, params, Engine.KEY_PARAM_SESSION_ID); 1898 copyStringParam(bundle, params, Engine.KEY_PARAM_UTTERANCE_ID); 1899 copyFloatParam(bundle, params, Engine.KEY_PARAM_VOLUME); 1900 copyFloatParam(bundle, params, Engine.KEY_PARAM_PAN); 1901 1902 // Copy feature strings defined by the framework. 1903 copyStringParam(bundle, params, Engine.KEY_FEATURE_NETWORK_SYNTHESIS); 1904 copyStringParam(bundle, params, Engine.KEY_FEATURE_EMBEDDED_SYNTHESIS); 1905 copyIntParam(bundle, params, Engine.KEY_FEATURE_NETWORK_TIMEOUT_MS); 1906 copyIntParam(bundle, params, Engine.KEY_FEATURE_NETWORK_RETRIES_COUNT); 1907 1908 // Copy over all parameters that start with the name of the 1909 // engine that we are currently connected to. The engine is 1910 // free to interpret them as it chooses. 1911 if (!TextUtils.isEmpty(mCurrentEngine)) { 1912 for (Map.Entry<String, String> entry : params.entrySet()) { 1913 final String key = entry.getKey(); 1914 if (key != null && key.startsWith(mCurrentEngine)) { 1915 bundle.putString(key, entry.getValue()); 1916 } 1917 } 1918 } 1919 1920 return bundle; 1921 } 1922 return null; 1923 } 1924 1925 private Bundle getParams(Bundle params) { 1926 if (params != null && !params.isEmpty()) { 1927 Bundle bundle = new Bundle(mParams); 1928 bundle.putAll(params); 1929 1930 verifyIntegerBundleParam(bundle, Engine.KEY_PARAM_STREAM); 1931 verifyIntegerBundleParam(bundle, Engine.KEY_PARAM_SESSION_ID); 1932 verifyStringBundleParam(bundle, Engine.KEY_PARAM_UTTERANCE_ID); 1933 verifyFloatBundleParam(bundle, Engine.KEY_PARAM_VOLUME); 1934 verifyFloatBundleParam(bundle, Engine.KEY_PARAM_PAN); 1935 1936 // Copy feature strings defined by the framework. 1937 verifyBooleanBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_SYNTHESIS); 1938 verifyBooleanBundleParam(bundle, Engine.KEY_FEATURE_EMBEDDED_SYNTHESIS); 1939 verifyIntegerBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_TIMEOUT_MS); 1940 verifyIntegerBundleParam(bundle, Engine.KEY_FEATURE_NETWORK_RETRIES_COUNT); 1941 1942 return bundle; 1943 } else { 1944 return mParams; 1945 } 1946 } 1947 1948 private static boolean verifyIntegerBundleParam(Bundle bundle, String key) { 1949 if (bundle.containsKey(key)) { 1950 if (!(bundle.get(key) instanceof Integer || 1951 bundle.get(key) instanceof Long)) { 1952 bundle.remove(key); 1953 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1954 + " with invalid type. Should be an Integer or a Long"); 1955 return false; 1956 } 1957 } 1958 return true; 1959 } 1960 1961 private static boolean verifyStringBundleParam(Bundle bundle, String key) { 1962 if (bundle.containsKey(key)) { 1963 if (!(bundle.get(key) instanceof String)) { 1964 bundle.remove(key); 1965 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1966 + " with invalid type. Should be a String"); 1967 return false; 1968 } 1969 } 1970 return true; 1971 } 1972 1973 private static boolean verifyBooleanBundleParam(Bundle bundle, String key) { 1974 if (bundle.containsKey(key)) { 1975 if (!(bundle.get(key) instanceof Boolean || 1976 bundle.get(key) instanceof String)) { 1977 bundle.remove(key); 1978 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1979 + " with invalid type. Should be a Boolean or String"); 1980 return false; 1981 } 1982 } 1983 return true; 1984 } 1985 1986 1987 private static boolean verifyFloatBundleParam(Bundle bundle, String key) { 1988 if (bundle.containsKey(key)) { 1989 if (!(bundle.get(key) instanceof Float || 1990 bundle.get(key) instanceof Double)) { 1991 bundle.remove(key); 1992 Log.w(TAG, "Synthesis request paramter " + key + " containst value " 1993 + " with invalid type. Should be a Float or a Double"); 1994 return false; 1995 } 1996 } 1997 return true; 1998 } 1999 2000 private void copyStringParam(Bundle bundle, HashMap<String, String> params, String key) { 2001 String value = params.get(key); 2002 if (value != null) { 2003 bundle.putString(key, value); 2004 } 2005 } 2006 2007 private void copyIntParam(Bundle bundle, HashMap<String, String> params, String key) { 2008 String valueString = params.get(key); 2009 if (!TextUtils.isEmpty(valueString)) { 2010 try { 2011 int value = Integer.parseInt(valueString); 2012 bundle.putInt(key, value); 2013 } catch (NumberFormatException ex) { 2014 // don't set the value in the bundle 2015 } 2016 } 2017 } 2018 2019 private void copyFloatParam(Bundle bundle, HashMap<String, String> params, String key) { 2020 String valueString = params.get(key); 2021 if (!TextUtils.isEmpty(valueString)) { 2022 try { 2023 float value = Float.parseFloat(valueString); 2024 bundle.putFloat(key, value); 2025 } catch (NumberFormatException ex) { 2026 // don't set the value in the bundle 2027 } 2028 } 2029 } 2030 2031 /** 2032 * Sets the listener that will be notified when synthesis of an utterance completes. 2033 * 2034 * @param listener The listener to use. 2035 * 2036 * @return {@link #ERROR} or {@link #SUCCESS}. 2037 * 2038 * @deprecated Use {@link #setOnUtteranceProgressListener(UtteranceProgressListener)} 2039 * instead. 2040 */ 2041 @Deprecated 2042 public int setOnUtteranceCompletedListener(final OnUtteranceCompletedListener listener) { 2043 mUtteranceProgressListener = UtteranceProgressListener.from(listener); 2044 return TextToSpeech.SUCCESS; 2045 } 2046 2047 /** 2048 * Sets the listener that will be notified of various events related to the 2049 * synthesis of a given utterance. 2050 * 2051 * See {@link UtteranceProgressListener} and 2052 * {@link TextToSpeech.Engine#KEY_PARAM_UTTERANCE_ID}. 2053 * 2054 * @param listener the listener to use. 2055 * @return {@link #ERROR} or {@link #SUCCESS} 2056 */ 2057 public int setOnUtteranceProgressListener(UtteranceProgressListener listener) { 2058 mUtteranceProgressListener = listener; 2059 return TextToSpeech.SUCCESS; 2060 } 2061 2062 /** 2063 * Sets the TTS engine to use. 2064 * 2065 * @deprecated This doesn't inform callers when the TTS engine has been 2066 * initialized. {@link #TextToSpeech(Context, OnInitListener, String)} 2067 * can be used with the appropriate engine name. Also, there is no 2068 * guarantee that the engine specified will be loaded. If it isn't 2069 * installed or disabled, the user / system wide defaults will apply. 2070 * 2071 * @param enginePackageName The package name for the synthesis engine (e.g. "com.svox.pico") 2072 * 2073 * @return {@link #ERROR} or {@link #SUCCESS}. 2074 */ 2075 @Deprecated 2076 public int setEngineByPackageName(String enginePackageName) { 2077 mRequestedEngine = enginePackageName; 2078 return initTts(); 2079 } 2080 2081 /** 2082 * Gets the package name of the default speech synthesis engine. 2083 * 2084 * @return Package name of the TTS engine that the user has chosen 2085 * as their default. 2086 */ 2087 public String getDefaultEngine() { 2088 return mEnginesHelper.getDefaultEngine(); 2089 } 2090 2091 /** 2092 * Checks whether the user's settings should override settings requested 2093 * by the calling application. As of the Ice cream sandwich release, 2094 * user settings never forcibly override the app's settings. 2095 */ 2096 @Deprecated 2097 public boolean areDefaultsEnforced() { 2098 return false; 2099 } 2100 2101 /** 2102 * Gets a list of all installed TTS engines. 2103 * 2104 * @return A list of engine info objects. The list can be empty, but never {@code null}. 2105 */ 2106 public List<EngineInfo> getEngines() { 2107 return mEnginesHelper.getEngines(); 2108 } 2109 2110 private class Connection implements ServiceConnection { 2111 private ITextToSpeechService mService; 2112 2113 private SetupConnectionAsyncTask mOnSetupConnectionAsyncTask; 2114 2115 private boolean mEstablished; 2116 2117 private final ITextToSpeechCallback.Stub mCallback = 2118 new ITextToSpeechCallback.Stub() { 2119 public void onStop(String utteranceId, boolean isStarted) 2120 throws RemoteException { 2121 UtteranceProgressListener listener = mUtteranceProgressListener; 2122 if (listener != null) { 2123 listener.onStop(utteranceId, isStarted); 2124 } 2125 }; 2126 2127 @Override 2128 public void onSuccess(String utteranceId) { 2129 UtteranceProgressListener listener = mUtteranceProgressListener; 2130 if (listener != null) { 2131 listener.onDone(utteranceId); 2132 } 2133 } 2134 2135 @Override 2136 public void onError(String utteranceId, int errorCode) { 2137 UtteranceProgressListener listener = mUtteranceProgressListener; 2138 if (listener != null) { 2139 listener.onError(utteranceId); 2140 } 2141 } 2142 2143 @Override 2144 public void onStart(String utteranceId) { 2145 UtteranceProgressListener listener = mUtteranceProgressListener; 2146 if (listener != null) { 2147 listener.onStart(utteranceId); 2148 } 2149 } 2150 2151 @Override 2152 public void onBeginSynthesis( 2153 String utteranceId, 2154 int sampleRateInHz, 2155 int audioFormat, 2156 int channelCount) { 2157 UtteranceProgressListener listener = mUtteranceProgressListener; 2158 if (listener != null) { 2159 listener.onBeginSynthesis( 2160 utteranceId, sampleRateInHz, audioFormat, channelCount); 2161 } 2162 } 2163 2164 @Override 2165 public void onAudioAvailable(String utteranceId, byte[] audio) { 2166 UtteranceProgressListener listener = mUtteranceProgressListener; 2167 if (listener != null) { 2168 listener.onAudioAvailable(utteranceId, audio); 2169 } 2170 } 2171 2172 @Override 2173 public void onRangeStart(String utteranceId, int start, int end, int frame) { 2174 UtteranceProgressListener listener = mUtteranceProgressListener; 2175 if (listener != null) { 2176 listener.onRangeStart(utteranceId, start, end, frame); 2177 } 2178 } 2179 }; 2180 2181 private class SetupConnectionAsyncTask extends AsyncTask<Void, Void, Integer> { 2182 private final ComponentName mName; 2183 2184 public SetupConnectionAsyncTask(ComponentName name) { 2185 mName = name; 2186 } 2187 2188 @Override 2189 protected Integer doInBackground(Void... params) { 2190 synchronized(mStartLock) { 2191 if (isCancelled()) { 2192 return null; 2193 } 2194 2195 try { 2196 mService.setCallback(getCallerIdentity(), mCallback); 2197 2198 if (mParams.getString(Engine.KEY_PARAM_LANGUAGE) == null) { 2199 String[] defaultLanguage = mService.getClientDefaultLanguage(); 2200 mParams.putString(Engine.KEY_PARAM_LANGUAGE, defaultLanguage[0]); 2201 mParams.putString(Engine.KEY_PARAM_COUNTRY, defaultLanguage[1]); 2202 mParams.putString(Engine.KEY_PARAM_VARIANT, defaultLanguage[2]); 2203 2204 // Get the default voice for the locale. 2205 String defaultVoiceName = mService.getDefaultVoiceNameFor( 2206 defaultLanguage[0], defaultLanguage[1], defaultLanguage[2]); 2207 mParams.putString(Engine.KEY_PARAM_VOICE_NAME, defaultVoiceName); 2208 } 2209 2210 Log.i(TAG, "Set up connection to " + mName); 2211 return SUCCESS; 2212 } catch (RemoteException re) { 2213 Log.e(TAG, "Error connecting to service, setCallback() failed"); 2214 return ERROR; 2215 } 2216 } 2217 } 2218 2219 @Override 2220 protected void onPostExecute(Integer result) { 2221 synchronized(mStartLock) { 2222 if (mOnSetupConnectionAsyncTask == this) { 2223 mOnSetupConnectionAsyncTask = null; 2224 } 2225 mEstablished = true; 2226 dispatchOnInit(result); 2227 } 2228 } 2229 } 2230 2231 @Override 2232 public void onServiceConnected(ComponentName name, IBinder service) { 2233 synchronized(mStartLock) { 2234 mConnectingServiceConnection = null; 2235 2236 Log.i(TAG, "Connected to " + name); 2237 2238 if (mOnSetupConnectionAsyncTask != null) { 2239 mOnSetupConnectionAsyncTask.cancel(false); 2240 } 2241 2242 mService = ITextToSpeechService.Stub.asInterface(service); 2243 mServiceConnection = Connection.this; 2244 2245 mEstablished = false; 2246 mOnSetupConnectionAsyncTask = new SetupConnectionAsyncTask(name); 2247 mOnSetupConnectionAsyncTask.execute(); 2248 } 2249 } 2250 2251 public IBinder getCallerIdentity() { 2252 return mCallback; 2253 } 2254 2255 /** 2256 * Clear connection related fields and cancel mOnServiceConnectedAsyncTask if set. 2257 * 2258 * @return true if we cancel mOnSetupConnectionAsyncTask in progress. 2259 */ 2260 private boolean clearServiceConnection() { 2261 synchronized(mStartLock) { 2262 boolean result = false; 2263 if (mOnSetupConnectionAsyncTask != null) { 2264 result = mOnSetupConnectionAsyncTask.cancel(false); 2265 mOnSetupConnectionAsyncTask = null; 2266 } 2267 2268 mService = null; 2269 // If this is the active connection, clear it 2270 if (mServiceConnection == this) { 2271 mServiceConnection = null; 2272 } 2273 return result; 2274 } 2275 } 2276 2277 @Override 2278 public void onServiceDisconnected(ComponentName name) { 2279 Log.i(TAG, "Asked to disconnect from " + name); 2280 if (clearServiceConnection()) { 2281 /* We need to protect against a rare case where engine 2282 * dies just after successful connection - and we process onServiceDisconnected 2283 * before OnServiceConnectedAsyncTask.onPostExecute. onServiceDisconnected cancels 2284 * OnServiceConnectedAsyncTask.onPostExecute and we don't call dispatchOnInit 2285 * with ERROR as argument. 2286 */ 2287 dispatchOnInit(ERROR); 2288 } 2289 } 2290 2291 public void disconnect() { 2292 mContext.unbindService(this); 2293 clearServiceConnection(); 2294 } 2295 2296 public boolean isEstablished() { 2297 return mService != null && mEstablished; 2298 } 2299 2300 public <R> R runAction(Action<R> action, R errorResult, String method, 2301 boolean reconnect, boolean onlyEstablishedConnection) { 2302 synchronized (mStartLock) { 2303 try { 2304 if (mService == null) { 2305 Log.w(TAG, method + " failed: not connected to TTS engine"); 2306 return errorResult; 2307 } 2308 if (onlyEstablishedConnection && !isEstablished()) { 2309 Log.w(TAG, method + " failed: TTS engine connection not fully set up"); 2310 return errorResult; 2311 } 2312 return action.run(mService); 2313 } catch (RemoteException ex) { 2314 Log.e(TAG, method + " failed", ex); 2315 if (reconnect) { 2316 disconnect(); 2317 initTts(); 2318 } 2319 return errorResult; 2320 } 2321 } 2322 } 2323 } 2324 2325 private interface Action<R> { 2326 R run(ITextToSpeechService service) throws RemoteException; 2327 } 2328 2329 /** 2330 * Information about an installed text-to-speech engine. 2331 * 2332 * @see TextToSpeech#getEngines 2333 */ 2334 public static class EngineInfo { 2335 /** 2336 * Engine package name.. 2337 */ 2338 public String name; 2339 /** 2340 * Localized label for the engine. 2341 */ 2342 public String label; 2343 /** 2344 * Icon for the engine. 2345 */ 2346 public int icon; 2347 /** 2348 * Whether this engine is a part of the system 2349 * image. 2350 * 2351 * @hide 2352 */ 2353 public boolean system; 2354 /** 2355 * The priority the engine declares for the the intent filter 2356 * {@code android.intent.action.TTS_SERVICE} 2357 * 2358 * @hide 2359 */ 2360 public int priority; 2361 2362 @Override 2363 public String toString() { 2364 return "EngineInfo{name=" + name + "}"; 2365 } 2366 2367 } 2368 2369 /** 2370 * Limit of length of input string passed to speak and synthesizeToFile. 2371 * 2372 * @see #speak 2373 * @see #synthesizeToFile 2374 */ 2375 public static int getMaxSpeechInputLength() { 2376 return 4000; 2377 } 2378 } 2379