Add support for voices in TTS API. (ad6df74a) · Commits · e / os / android_frameworks_base

api/current.txt

+42 −6

Original line number	Diff line number	Diff line
		@@ -26609,6 +26609,7 @@ package android.speech.tts {
		method public int getSpeechRate();
		method public deprecated java.lang.String getText();
		method public java.lang.String getVariant();
		method public java.lang.String getVoiceName();
		}

		public class TextToSpeech {
		@@ -26620,13 +26621,17 @@ package android.speech.tts {
		method public int addSpeech(java.lang.CharSequence, java.lang.String, int);
		method public int addSpeech(java.lang.String, java.lang.String);
		method public int addSpeech(java.lang.CharSequence, java.lang.String);
		method public boolean areDefaultsEnforced();
		method public deprecated boolean areDefaultsEnforced();
		method public java.util.Set<java.util.Locale> getAvailableLanguages();
		method public java.lang.String getDefaultEngine();
		method public java.util.Locale getDefaultLanguage();
		method public deprecated java.util.Locale getDefaultLanguage();
		method public android.speech.tts.Voice getDefaultVoice();
		method public java.util.List<android.speech.tts.TextToSpeech.EngineInfo> getEngines();
		method public java.util.Set<java.lang.String> getFeatures(java.util.Locale);
		method public java.util.Locale getLanguage();
		method public deprecated java.util.Set<java.lang.String> getFeatures(java.util.Locale);
		method public deprecated java.util.Locale getLanguage();
		method public static int getMaxSpeechInputLength();
		method public android.speech.tts.Voice getVoice();
		method public java.util.Set<android.speech.tts.Voice> getVoices();
		method public int isLanguageAvailable(java.util.Locale);
		method public boolean isSpeaking();
		method public int playEarcon(java.lang.String, int, java.util.HashMap<java.lang.String, java.lang.String>, java.lang.String);
		@@ -26639,6 +26644,7 @@ package android.speech.tts {
		method public int setOnUtteranceProgressListener(android.speech.tts.UtteranceProgressListener);
		method public int setPitch(float);
		method public int setSpeechRate(float);
		method public int setVoice(android.speech.tts.Voice);
		method public void shutdown();
		method public int speak(java.lang.CharSequence, int, java.util.HashMap<java.lang.String, java.lang.String>, java.lang.String);
		method public deprecated int speak(java.lang.String, int, java.util.HashMap<java.lang.String, java.lang.String>);
		@@ -26650,6 +26656,7 @@ package android.speech.tts {
		field public static final int ERROR_INVALID_REQUEST = -8; // 0xfffffff8
		field public static final int ERROR_NETWORK = -6; // 0xfffffffa
		field public static final int ERROR_NETWORK_TIMEOUT = -7; // 0xfffffff9
		field public static final int ERROR_NOT_INSTALLED_YET = -9; // 0xfffffff7
		field public static final int ERROR_OUTPUT = -5; // 0xfffffffb
		field public static final int ERROR_SERVICE = -4; // 0xfffffffc
		field public static final int ERROR_SYNTHESIS = -3; // 0xfffffffd
		@@ -26685,8 +26692,11 @@ package android.speech.tts {
		field public static final deprecated java.lang.String EXTRA_VOICE_DATA_FILES_INFO = "dataFilesInfo";
		field public static final deprecated java.lang.String EXTRA_VOICE_DATA_ROOT_DIRECTORY = "dataRoot";
		field public static final java.lang.String INTENT_ACTION_TTS_SERVICE = "android.intent.action.TTS_SERVICE";
		field public static final java.lang.String KEY_FEATURE_EMBEDDED_SYNTHESIS = "embeddedTts";
		field public static final java.lang.String KEY_FEATURE_NETWORK_SYNTHESIS = "networkTts";
		field public static final deprecated java.lang.String KEY_FEATURE_EMBEDDED_SYNTHESIS = "embeddedTts";
		field public static final java.lang.String KEY_FEATURE_NETWORK_RETRIES_COUNT = "networkRetriesCount";
		field public static final deprecated java.lang.String KEY_FEATURE_NETWORK_SYNTHESIS = "networkTts";
		field public static final java.lang.String KEY_FEATURE_NETWORK_TIMEOUT_MS = "networkTimeoutMs";
		field public static final java.lang.String KEY_FEATURE_NOT_INSTALLED = "notInstalled";
		field public static final java.lang.String KEY_PARAM_PAN = "pan";
		field public static final java.lang.String KEY_PARAM_SESSION_ID = "sessionId";
		field public static final java.lang.String KEY_PARAM_STREAM = "streamType";
		@@ -26712,11 +26722,15 @@ package android.speech.tts {

		public abstract class TextToSpeechService extends android.app.Service {
		ctor public TextToSpeechService();
		method protected int isValidVoiceName(java.lang.String);
		method public android.os.IBinder onBind(android.content.Intent);
		method protected java.lang.String onGetDefaultVoiceNameFor(java.lang.String, java.lang.String, java.lang.String);
		method protected java.util.Set<java.lang.String> onGetFeaturesForLanguage(java.lang.String, java.lang.String, java.lang.String);
		method protected abstract java.lang.String[] onGetLanguage();
		method protected java.util.List<android.speech.tts.Voice> onGetVoices();
		method protected abstract int onIsLanguageAvailable(java.lang.String, java.lang.String, java.lang.String);
		method protected abstract int onLoadLanguage(java.lang.String, java.lang.String, java.lang.String);
		method protected int onLoadVoice(java.lang.String);
		method protected abstract void onStop();
		method protected abstract void onSynthesizeText(android.speech.tts.SynthesisRequest, android.speech.tts.SynthesisCallback);
		}
		@@ -26808,6 +26822,28 @@ package android.speech.tts {
		method public abstract void onStart(java.lang.String);
		}

		public class Voice implements android.os.Parcelable {
		ctor public Voice(java.lang.String, java.util.Locale, int, int, boolean, java.util.Set<java.lang.String>);
		method public int describeContents();
		method public java.util.Set<java.lang.String> getFeatures();
		method public int getLatency();
		method public java.util.Locale getLocale();
		method public java.lang.String getName();
		method public int getQuality();
		method public boolean getRequiresNetworkConnection();
		method public void writeToParcel(android.os.Parcel, int);
		field public static final int LATENCY_HIGH = 400; // 0x190
		field public static final int LATENCY_LOW = 200; // 0xc8
		field public static final int LATENCY_NORMAL = 300; // 0x12c
		field public static final int LATENCY_VERY_HIGH = 500; // 0x1f4
		field public static final int LATENCY_VERY_LOW = 100; // 0x64
		field public static final int QUALITY_HIGH = 400; // 0x190
		field public static final int QUALITY_LOW = 200; // 0xc8
		field public static final int QUALITY_NORMAL = 300; // 0x12c
		field public static final int QUALITY_VERY_HIGH = 500; // 0x1f4
		field public static final int QUALITY_VERY_LOW = 100; // 0x64
		}

		}

		package android.system {

core/java/android/speech/tts/ITextToSpeechService.aidl

+34 −0

Original line number	Diff line number	Diff line
		@@ -20,6 +20,7 @@ import android.net.Uri;
		import android.os.Bundle;
		import android.os.ParcelFileDescriptor;
		import android.speech.tts.ITextToSpeechCallback;
		import android.speech.tts.Voice;

		/**
		* Interface for TextToSpeech to talk to TextToSpeechService.
		@@ -173,4 +174,37 @@ interface ITextToSpeechService {
		* @param cb The callback.
		*/
		void setCallback(in IBinder caller, ITextToSpeechCallback cb);

		/**
		* Get the array of available voices.
		*/
		List<Voice> getVoices();

		/**
		* Notifies the engine that it should load a speech synthesis voice.
		*
		* @param caller a binder representing the identity of the calling
		* TextToSpeech object.
		* @param voiceName Unique voice of the name.
		* @return {@link TextToSpeech#SUCCESS} or {@link TextToSpeech#ERROR}.
		*/
		int loadVoice(in IBinder caller, in String voiceName);

		/**
		* Return a name of the default voice for a given locale.
		*
		* This allows {@link TextToSpeech#getVoice} to return a sensible value after a client calls
		* {@link TextToSpeech#setLanguage}.
		*
		* @param lang ISO 3-character language code.
		* @param country ISO 3-character country code. May be empty or null.
		* @param variant Language variant. May be empty or null.
		* @return Code indicating the support status for the locale.
		* One of {@link TextToSpeech#LANG_AVAILABLE},
		* {@link TextToSpeech#LANG_COUNTRY_AVAILABLE},
		* {@link TextToSpeech#LANG_COUNTRY_VAR_AVAILABLE},
		* {@link TextToSpeech#LANG_MISSING_DATA}
		* {@link TextToSpeech#LANG_NOT_SUPPORTED}.
		*/
		String getDefaultVoiceNameFor(in String lang, in String country, in String variant);
		}

core/java/android/speech/tts/SynthesisRequest.java

+19 −1

Original line number	Diff line number	Diff line
		@@ -24,6 +24,9 @@ import android.os.Bundle;
		* <li>The synthesis locale, represented as a language, country and a variant.
		* The language is an ISO 639-3 letter language code, and the country is an
		* ISO 3166 alpha 3 code. The variant is not specified.</li>
		* <li>The name of the voice requested for this synthesis. May be empty if
		* the client uses {@link TextToSpeech#setLanguage} instead of
		* {@link TextToSpeech#setVoice}</li>
		* <li>The synthesis speech rate, with 100 being the normal, and
		* higher values representing higher speech rates.</li>
		* <li>The voice pitch, with 100 being the default pitch.</li>
		@@ -36,6 +39,7 @@ import android.os.Bundle;
		public final class SynthesisRequest {
		private final CharSequence mText;
		private final Bundle mParams;
		private String mVoiceName;
		private String mLanguage;
		private String mCountry;
		private String mVariant;
		@@ -71,6 +75,13 @@ public final class SynthesisRequest {
		return mText;
		}

		/**
		* Gets the name of the voice to use.
		*/
		public String getVoiceName() {
		return mVoiceName;
		}

		/**
		* Gets the ISO 3-letter language code for the language to use.
		*/
		@@ -129,6 +140,13 @@ public final class SynthesisRequest {
		mVariant = variant;
		}

		/**
		* Sets the voice name for the request.
		*/
		void setVoiceName(String voiceName) {
		mVoiceName = voiceName;
		}

		/**
		* Sets the speech rate.
		*/

core/java/android/speech/tts/TextToSpeech.java

+283 −13

File changed.

Preview size limit exceeded, changes collapsed.

core/java/android/speech/tts/TextToSpeechService.java

+254 −1

Original line number	Diff line number	Diff line
		@@ -39,6 +39,7 @@ import android.util.Log;
		import java.io.FileOutputStream;
		import java.io.IOException;
		import java.util.ArrayList;
		import java.util.Collections;
		import java.util.HashMap;
		import java.util.List;
		import java.util.Locale;
		@@ -49,7 +50,7 @@ import java.util.Set;

		/**
		* Abstract base class for TTS engine implementations. The following methods
		* need to be implemented for V1 API ({@link TextToSpeech}) implementation.
		* need to be implemented:
		* <ul>
		* <li>{@link #onIsLanguageAvailable}</li>
		* <li>{@link #onLoadLanguage}</li>
		@@ -76,6 +77,29 @@ import java.util.Set;
		*
		* {@link #onGetLanguage} is not required as of JELLYBEAN_MR2 (API 18) and later, it is only
		* called on earlier versions of Android.
		*
		* API Level 20 adds support for Voice objects. Voices are an abstraction that allow the TTS
		* service to expose multiple backends for a single locale. Each one of them can have a different
		* features set. In order to fully take advantage of voices, an engine should implement
		* the following methods:
		* <ul>
		* <li>{@link #onGetVoices()}</li>
		* <li>{@link #isValidVoiceName(String)}</li>
		* <li>{@link #onLoadVoice(String)}</li>
		* <li>{@link #onGetDefaultVoiceNameFor(String, String, String)}</li>
		* </ul>
		* The first three methods are siblings of the {@link #onGetLanguage},
		* {@link #onIsLanguageAvailable} and {@link #onLoadLanguage} methods. The last one,
		* {@link #onGetDefaultVoiceNameFor(String, String, String)} is a link between locale and voice
		* based methods. Since API level 20 {@link TextToSpeech#setLanguage} is implemented by
		* calling {@link TextToSpeech#setVoice} with the voice returned by
		* {@link #onGetDefaultVoiceNameFor(String, String, String)}.
		*
		* If the client uses a voice instead of a locale, {@link SynthesisRequest} will contain the
		* requested voice name.
		*
		* The default implementations of Voice-related methods implement them using the
		* pre-existing locale-based implementation.
		*/
		public abstract class TextToSpeechService extends Service {

		@@ -228,6 +252,160 @@ public abstract class TextToSpeechService extends Service {
		return null;
		}

		private int getExpectedLanguageAvailableStatus(Locale locale) {
		int expectedStatus = TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE;
		if (locale.getVariant().isEmpty()) {
		if (locale.getCountry().isEmpty()) {
		expectedStatus = TextToSpeech.LANG_AVAILABLE;
		} else {
		expectedStatus = TextToSpeech.LANG_COUNTRY_AVAILABLE;
		}
		}
		return expectedStatus;
		}

		/**
		* Queries the service for a set of supported voices.
		*
		* Can be called on multiple threads.
		*
		* The default implementation tries to enumerate all available locales, pass them to
		* {@link #onIsLanguageAvailable(String, String, String)} and create Voice instances (using
		* the locale's BCP-47 language tag as the voice name) for the ones that are supported.
		* Note, that this implementation is suitable only for engines that don't have multiple voices
		* for a single locale. Also, this implementation won't work with Locales not listed in the
		* set returned by the {@link Locale#getAvailableLocales()} method.
		*
		* @return A list of voices supported.
		*/
		protected List<Voice> onGetVoices() {
		// Enumerate all locales and check if they are available
		ArrayList<Voice> voices = new ArrayList<Voice>();
		for (Locale locale : Locale.getAvailableLocales()) {
		int expectedStatus = getExpectedLanguageAvailableStatus(locale);
		try {
		int localeStatus = onIsLanguageAvailable(locale.getISO3Language(),
		locale.getISO3Country(), locale.getVariant());
		if (localeStatus != expectedStatus) {
		continue;
		}
		} catch (MissingResourceException e) {
		// Ignore locale without iso 3 codes
		continue;
		}
		Set<String> features = onGetFeaturesForLanguage(locale.getISO3Language(),
		locale.getISO3Country(), locale.getVariant());
		voices.add(new Voice(locale.toLanguageTag(), locale, Voice.QUALITY_NORMAL,
		Voice.LATENCY_NORMAL, false, features));
		}
		return voices;
		}

		/**
		* Return a name of the default voice for a given locale.
		*
		* This method provides a mapping between locales and available voices. This method is
		* used in {@link TextToSpeech#setLanguage}, which calls this method and then calls
		* {@link TextToSpeech#setVoice} with the voice returned by this method.
		*
		* Also, it's used by {@link TextToSpeech#getDefaultVoice()} to find a default voice for
		* the default locale.
		*
		* @param lang ISO-3 language code.
		* @param country ISO-3 country code. May be empty or null.
		* @param variant Language variant. May be empty or null.

		* @return A name of the default voice for a given locale.
		*/
		protected String onGetDefaultVoiceNameFor(String lang, String country, String variant) {
		int localeStatus = onIsLanguageAvailable(lang, country, variant);
		Locale iso3Locale = null;
		switch (localeStatus) {
		case TextToSpeech.LANG_AVAILABLE:
		iso3Locale = new Locale(lang);
		break;
		case TextToSpeech.LANG_COUNTRY_AVAILABLE:
		iso3Locale = new Locale(lang, country);
		break;
		case TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE:
		iso3Locale = new Locale(lang, country, variant);
		break;
		default:
		return null;
		}
		Locale properLocale = TtsEngines.normalizeTTSLocale(iso3Locale);
		String voiceName = properLocale.toLanguageTag();
		if (isValidVoiceName(voiceName) == TextToSpeech.SUCCESS) {
		return voiceName;
		} else {
		return null;
		}
		}

		/**
		* Notifies the engine that it should load a speech synthesis voice. There is no guarantee
		* that this method is always called before the voice is used for synthesis. It is merely
		* a hint to the engine that it will probably get some synthesis requests for this voice
		* at some point in the future.
		*
		* Will be called only on synthesis thread.
		*
		* The default implementation creates a Locale from the voice name (by interpreting the name as
		* a BCP-47 tag for the locale), and passes it to
		* {@link #onLoadLanguage(String, String, String)}.
		*
		* @param voiceName Name of the voice.
		* @return {@link TextToSpeech#ERROR} or {@link TextToSpeech#SUCCESS}.
		*/
		protected int onLoadVoice(String voiceName) {
		Locale locale = Locale.forLanguageTag(voiceName);
		if (locale == null) {
		return TextToSpeech.ERROR;
		}
		int expectedStatus = getExpectedLanguageAvailableStatus(locale);
		try {
		int localeStatus = onIsLanguageAvailable(locale.getISO3Language(),
		locale.getISO3Country(), locale.getVariant());
		if (localeStatus != expectedStatus) {
		return TextToSpeech.ERROR;
		}
		onLoadLanguage(locale.getISO3Language(),
		locale.getISO3Country(), locale.getVariant());
		return TextToSpeech.SUCCESS;
		} catch (MissingResourceException e) {
		return TextToSpeech.ERROR;
		}
		}

		/**
		* Checks whether the engine supports a voice with a given name.
		*
		* Can be called on multiple threads.
		*
		* The default implementation treats the voice name as a language tag, creating a Locale from
		* the voice name, and passes it to {@link #onIsLanguageAvailable(String, String, String)}.
		*
		* @param voiceName Name of the voice.
		* @return {@link TextToSpeech#ERROR} or {@link TextToSpeech#SUCCESS}.
		*/
		protected int isValidVoiceName(String voiceName) {
		Locale locale = Locale.forLanguageTag(voiceName);
		if (locale == null) {
		return TextToSpeech.ERROR;
		}
		int expectedStatus = getExpectedLanguageAvailableStatus(locale);
		try {
		int localeStatus = onIsLanguageAvailable(locale.getISO3Language(),
		locale.getISO3Country(), locale.getVariant());
		if (localeStatus != expectedStatus) {
		return TextToSpeech.ERROR;
		}
		return TextToSpeech.SUCCESS;
		} catch (MissingResourceException e) {
		return TextToSpeech.ERROR;
		}
		}

		private int getDefaultSpeechRate() {
		return getSecureSettingInt(Settings.Secure.TTS_DEFAULT_RATE, Engine.DEFAULT_RATE);
		}
		@@ -736,7 +914,11 @@ public abstract class TextToSpeechService extends Service {
		}

		private void setRequestParams(SynthesisRequest request) {
		String voiceName = getVoiceName();
		request.setLanguage(getLanguage(), getCountry(), getVariant());
		if (!TextUtils.isEmpty(voiceName)) {
		request.setVoiceName(getVoiceName());
		}
		request.setSpeechRate(getSpeechRate());
		request.setCallerUid(mCallerUid);
		request.setPitch(getPitch());
		@@ -770,6 +952,10 @@ public abstract class TextToSpeechService extends Service {
		public String getLanguage() {
		return getStringParam(mParams, Engine.KEY_PARAM_LANGUAGE, mDefaultLocale[0]);
		}

		public String getVoiceName() {
		return getStringParam(mParams, Engine.KEY_PARAM_VOICE_NAME, "");
		}
		}

		private class SynthesisToFileOutputStreamSpeechItemV1 extends SynthesisSpeechItemV1 {
		@@ -896,6 +1082,35 @@ public abstract class TextToSpeechService extends Service {
		}
		}

		/**
		* Call {@link TextToSpeechService#onLoadLanguage} on synth thread.
		*/
		private class LoadVoiceItem extends SpeechItem {
		private final String mVoiceName;

		public LoadVoiceItem(Object callerIdentity, int callerUid, int callerPid,
		String voiceName) {
		super(callerIdentity, callerUid, callerPid);
		mVoiceName = voiceName;
		}

		@Override
		public boolean isValid() {
		return true;
		}

		@Override
		protected void playImpl() {
		TextToSpeechService.this.onLoadVoice(mVoiceName);
		}

		@Override
		protected void stopImpl() {
		// No-op
		}
		}


		@Override
		public IBinder onBind(Intent intent) {
		if (TextToSpeech.Engine.INTENT_ACTION_TTS_SERVICE.equals(intent.getAction())) {
		@@ -1041,6 +1256,44 @@ public abstract class TextToSpeechService extends Service {
		return retVal;
		}

		@Override
		public List<Voice> getVoices() {
		return onGetVoices();
		}

		@Override
		public int loadVoice(IBinder caller, String voiceName) {
		if (!checkNonNull(voiceName)) {
		return TextToSpeech.ERROR;
		}
		int retVal = isValidVoiceName(voiceName);

		if (retVal == TextToSpeech.SUCCESS) {
		SpeechItem item = new LoadVoiceItem(caller, Binder.getCallingUid(),
		Binder.getCallingPid(), voiceName);
		if (mSynthHandler.enqueueSpeechItem(TextToSpeech.QUEUE_ADD, item) !=
		TextToSpeech.SUCCESS) {
		return TextToSpeech.ERROR;
		}
		}
		return retVal;
		}

		public String getDefaultVoiceNameFor(String lang, String country, String variant) {
		if (!checkNonNull(lang)) {
		return null;
		}
		int retVal = onIsLanguageAvailable(lang, country, variant);

		if (retVal == TextToSpeech.LANG_AVAILABLE \|\|
		retVal == TextToSpeech.LANG_COUNTRY_AVAILABLE \|\|
		retVal == TextToSpeech.LANG_COUNTRY_VAR_AVAILABLE) {
		return onGetDefaultVoiceNameFor(lang, country, variant);
		} else {
		return null;
		}
		}

		@Override
		public void setCallback(IBinder caller, ITextToSpeechCallback cb) {
		// Note that passing in a null callback is a valid use case.