Markup support for framework (a6cc9b85) · Commits · e / os / android_frameworks_base

api/current.txt

+105 −1

Original line number	Diff line number	Diff line
		@@ -26581,6 +26581,28 @@ package android.speech {

		package android.speech.tts {

		public final class Markup implements android.os.Parcelable {
		ctor public Markup();
		ctor public Markup(java.lang.String);
		ctor public Markup(android.speech.tts.Markup);
		method public android.speech.tts.Markup addNestedMarkup(android.speech.tts.Markup);
		method public int describeContents();
		method public android.speech.tts.Markup getNestedMarkup(int);
		method public java.util.List<android.speech.tts.Markup> getNestedMarkups();
		method public java.lang.String getParameter(java.lang.String);
		method public java.lang.String getPlainText();
		method public java.lang.String getType();
		method public static android.speech.tts.Markup markupFromString(java.lang.String) throws java.lang.IllegalArgumentException;
		method public int nestedMarkupSize();
		method public int parametersSize();
		method public boolean removeNestedMarkup(android.speech.tts.Markup);
		method public void removeParameter(java.lang.String);
		method public android.speech.tts.Markup setParameter(java.lang.String, java.lang.String);
		method public void setPlainText(java.lang.String);
		method public void setType(java.lang.String);
		method public void writeToParcel(android.os.Parcel, int);
		}

		public final class RequestConfig {
		method public android.os.Bundle getAudioParams();
		method public android.speech.tts.VoiceInfo getVoice();
		@@ -26643,9 +26665,10 @@ package android.speech.tts {
		}

		public final class SynthesisRequestV2 implements android.os.Parcelable {
		ctor public SynthesisRequestV2(java.lang.String, java.lang.String, java.lang.String, android.os.Bundle, android.os.Bundle);
		ctor public SynthesisRequestV2(android.speech.tts.Markup, java.lang.String, java.lang.String, android.os.Bundle, android.os.Bundle);
		method public int describeContents();
		method public android.os.Bundle getAudioParams();
		method public android.speech.tts.Markup getMarkup();
		method public java.lang.String getText();
		method public java.lang.String getUtteranceId();
		method public java.lang.String getVoiceName();
		@@ -26748,7 +26771,9 @@ package android.speech.tts {
		method public void queueAudio(android.net.Uri, android.speech.tts.TextToSpeechClient.UtteranceId, android.speech.tts.RequestConfig, android.speech.tts.TextToSpeechClient.RequestCallbacks);
		method public void queueSilence(long, android.speech.tts.TextToSpeechClient.UtteranceId, android.speech.tts.TextToSpeechClient.RequestCallbacks);
		method public void queueSpeak(java.lang.String, android.speech.tts.TextToSpeechClient.UtteranceId, android.speech.tts.RequestConfig, android.speech.tts.TextToSpeechClient.RequestCallbacks);
		method public void queueSpeak(android.speech.tts.Markup, android.speech.tts.TextToSpeechClient.UtteranceId, android.speech.tts.RequestConfig, android.speech.tts.TextToSpeechClient.RequestCallbacks);
		method public void queueSynthesizeToFile(java.lang.String, android.speech.tts.TextToSpeechClient.UtteranceId, java.io.File, android.speech.tts.RequestConfig, android.speech.tts.TextToSpeechClient.RequestCallbacks);
		method public void queueSynthesizeToFile(android.speech.tts.Markup, android.speech.tts.TextToSpeechClient.UtteranceId, java.io.File, android.speech.tts.RequestConfig, android.speech.tts.TextToSpeechClient.RequestCallbacks);
		method public void stop();
		}

		@@ -26822,6 +26847,85 @@ package android.speech.tts {
		method protected void onVoicesInfoChange();
		}

		public class Utterance {
		ctor public Utterance();
		method public android.speech.tts.Utterance append(android.speech.tts.Utterance.AbstractTts<? extends android.speech.tts.Utterance.AbstractTts<?>>);
		method public android.speech.tts.Utterance append(java.lang.String);
		method public android.speech.tts.Utterance append(int);
		method public android.speech.tts.Markup createMarkup();
		method public android.speech.tts.Utterance.AbstractTts<? extends android.speech.tts.Utterance.AbstractTts<?>> get(int);
		method public android.speech.tts.Utterance setNoWarningOnFallback(boolean);
		method public int size();
		method public static android.speech.tts.Utterance utteranceFromString(java.lang.String) throws java.lang.IllegalArgumentException;
		field public static final int ANIMACY_ANIMATE = 1; // 0x1
		field public static final int ANIMACY_INANIMATE = 2; // 0x2
		field public static final int ANIMACY_UNKNOWN = 0; // 0x0
		field public static final int CASE_ABLATIVE = 4; // 0x4
		field public static final int CASE_ACCUSATIVE = 2; // 0x2
		field public static final int CASE_DATIVE = 3; // 0x3
		field public static final int CASE_GENITIVE = 5; // 0x5
		field public static final int CASE_INSTRUMENTAL = 8; // 0x8
		field public static final int CASE_LOCATIVE = 7; // 0x7
		field public static final int CASE_NOMINATIVE = 1; // 0x1
		field public static final int CASE_UNKNOWN = 0; // 0x0
		field public static final int CASE_VOCATIVE = 6; // 0x6
		field public static final int GENDER_FEMALE = 3; // 0x3
		field public static final int GENDER_MALE = 2; // 0x2
		field public static final int GENDER_NEUTRAL = 1; // 0x1
		field public static final int GENDER_UNKNOWN = 0; // 0x0
		field public static final java.lang.String KEY_NO_WARNING_ON_FALLBACK = "no_warning_on_fallback";
		field public static final int MULTIPLICITY_DUAL = 2; // 0x2
		field public static final int MULTIPLICITY_PLURAL = 3; // 0x3
		field public static final int MULTIPLICITY_SINGLE = 1; // 0x1
		field public static final int MULTIPLICITY_UNKNOWN = 0; // 0x0
		field public static final java.lang.String TYPE_UTTERANCE = "utterance";
		}

		public static abstract class Utterance.AbstractTts {
		ctor protected Utterance.AbstractTts();
		ctor protected Utterance.AbstractTts(android.speech.tts.Markup);
		method public java.lang.String generatePlainText();
		method public android.speech.tts.Markup getMarkup();
		method protected java.lang.String getParameter(java.lang.String);
		method public java.lang.String getPlainText();
		method public java.lang.String getType();
		method protected C removeParameter(java.lang.String);
		method protected C setParameter(java.lang.String, java.lang.String);
		method public C setPlainText(java.lang.String);
		field protected android.speech.tts.Markup mMarkup;
		}

		public static abstract class Utterance.AbstractTtsSemioticClass extends android.speech.tts.Utterance.AbstractTts {
		ctor protected Utterance.AbstractTtsSemioticClass();
		ctor protected Utterance.AbstractTtsSemioticClass(android.speech.tts.Markup);
		method public int getAnimacy();
		method public int getCase();
		method public int getGender();
		method public int getMultiplicity();
		method public C setAnimacy(int);
		method public C setCase(int);
		method public C setGender(int);
		method public C setMultiplicity(int);
		}

		public static class Utterance.TtsCardinal extends android.speech.tts.Utterance.AbstractTtsSemioticClass {
		ctor public Utterance.TtsCardinal();
		ctor public Utterance.TtsCardinal(int);
		ctor public Utterance.TtsCardinal(java.lang.String);
		method public java.lang.String getInteger();
		method public android.speech.tts.Utterance.TtsCardinal setInteger(int);
		method public android.speech.tts.Utterance.TtsCardinal setInteger(java.lang.String);
		field protected static final java.lang.String TYPE_CARDINAL = "cardinal";
		}

		public static class Utterance.TtsText extends android.speech.tts.Utterance.AbstractTtsSemioticClass {
		ctor public Utterance.TtsText();
		ctor public Utterance.TtsText(java.lang.String);
		method public java.lang.String getText();
		method public android.speech.tts.Utterance.TtsText setText(java.lang.String);
		field protected static final java.lang.String TYPE_TEXT = "text";
		}

		public abstract class UtteranceProgressListener {
		ctor public UtteranceProgressListener();
		method public abstract void onDone(java.lang.String);

core/java/android/speech/tts/Markup.java

0 → 100644

+537 −0

File added.

Preview size limit exceeded, changes collapsed.

core/java/android/speech/tts/SynthesisRequestV2.java

+28 −10

Original line number	Diff line number	Diff line
		@@ -4,11 +4,12 @@ import android.os.Bundle;
		import android.os.Parcel;
		import android.os.Parcelable;
		import android.speech.tts.TextToSpeechClient.UtteranceId;
		import android.util.Log;

		/**
		* Service-side representation of a synthesis request from a V2 API client. Contains:
		* <ul>
		* <li>The utterance to synthesize</li>
		* <li>The markup object to synthesize containing the utterance.</li>
		* <li>The id of the utterance (String, result of {@link UtteranceId#toUniqueString()}</li>
		* <li>The synthesis voice name (String, result of {@link VoiceInfo#getName()})</li>
		* <li>Voice parameters (Bundle of parameters)</li>
		@@ -16,8 +17,11 @@ import android.speech.tts.TextToSpeechClient.UtteranceId;
		* </ul>
		*/
		public final class SynthesisRequestV2 implements Parcelable {
		/** Synthesis utterance. */
		private final String mText;

		private static final String TAG = "SynthesisRequestV2";

		/** Synthesis markup */
		private final Markup mMarkup;

		/** Synthesis id. */
		private final String mUtteranceId;
		@@ -34,9 +38,9 @@ public final class SynthesisRequestV2 implements Parcelable {
		/**
		* Constructor for test purposes.
		*/
		public SynthesisRequestV2(String text, String utteranceId, String voiceName,
		public SynthesisRequestV2(Markup markup, String utteranceId, String voiceName,
		Bundle voiceParams, Bundle audioParams) {
		this.mText = text;
		this.mMarkup = markup;
		this.mUtteranceId = utteranceId;
		this.mVoiceName = voiceName;
		this.mVoiceParams = voiceParams;
		@@ -49,15 +53,18 @@ public final class SynthesisRequestV2 implements Parcelable {
		* @hide
		*/
		public SynthesisRequestV2(Parcel in) {
		this.mText = in.readString();
		this.mMarkup = (Markup) in.readValue(Markup.class.getClassLoader());
		this.mUtteranceId = in.readString();
		this.mVoiceName = in.readString();
		this.mVoiceParams = in.readBundle();
		this.mAudioParams = in.readBundle();
		}

		SynthesisRequestV2(String text, String utteranceId, RequestConfig rconfig) {
		this.mText = text;
		/**
		* Constructor to request the synthesis of a sentence.
		*/
		SynthesisRequestV2(Markup markup, String utteranceId, RequestConfig rconfig) {
		this.mMarkup = markup;
		this.mUtteranceId = utteranceId;
		this.mVoiceName = rconfig.getVoice().getName();
		this.mVoiceParams = rconfig.getVoiceParams();
		@@ -71,7 +78,7 @@ public final class SynthesisRequestV2 implements Parcelable {
		*/
		@Override
		public void writeToParcel(Parcel dest, int flags) {
		dest.writeString(mText);
		dest.writeValue(mMarkup);
		dest.writeString(mUtteranceId);
		dest.writeString(mVoiceName);
		dest.writeBundle(mVoiceParams);
		@@ -82,7 +89,18 @@ public final class SynthesisRequestV2 implements Parcelable {
		* @return the text which should be synthesized.
		*/
		public String getText() {
		return mText;
		if (mMarkup.getPlainText() == null) {
		Log.e(TAG, "Plaintext of markup is null.");
		return "";
		}
		return mMarkup.getPlainText();
		}

		/**
		* @return the markup which should be synthesized.
		*/
		public Markup getMarkup() {
		return mMarkup;
		}

		/**

core/java/android/speech/tts/TextToSpeechClient.java

+60 −7

Original line number	Diff line number	Diff line
		@@ -512,7 +512,6 @@ public class TextToSpeechClient {
		}
		}


		/**
		* Connects the client to TTS service. This method returns immediately, and connects to the
		* service in the background.
		@@ -876,7 +875,7 @@ public class TextToSpeechClient {
		private static final String ACTION_QUEUE_SPEAK_NAME = "queueSpeak";

		/**
		* Speaks the string using the specified queuing strategy using current
		* Speaks the string using the specified queuing strategy and the current
		* voice. This method is asynchronous, i.e. the method just adds the request
		* to the queue of TTS requests and then returns. The synthesis might not
		* have finished (or even started!) at the time when this method returns.
		@@ -887,12 +886,35 @@ public class TextToSpeechClient {
		* in {@link RequestCallbacks}.
		* @param config Synthesis request configuration. Can't be null. Has to contain a
		* voice.
		* @param callbacks Synthesis request callbacks. If null, default request
		* @param callbacks Synthesis request callbacks. If null, the default request
		* callbacks object will be used.
		*/
		public void queueSpeak(final String utterance, final UtteranceId utteranceId,
		final RequestConfig config,
		final RequestCallbacks callbacks) {
		queueSpeak(createMarkupFromString(utterance), utteranceId, config, callbacks);
		}

		/**
		* Speaks the {@link Markup} (which can be constructed with {@link Utterance}) using
		* the specified queuing strategy and the current voice. This method is
		* asynchronous, i.e. the method just adds the request to the queue of TTS
		* requests and then returns. The synthesis might not have finished (or even
		* started!) at the time when this method returns.
		*
		* @param markup The Markup to be spoken. The written equivalent of the spoken
		* text should be no longer than 1000 characters.
		* @param utteranceId Unique identificator used to track the synthesis progress
		* in {@link RequestCallbacks}.
		* @param config Synthesis request configuration. Can't be null. Has to contain a
		* voice.
		* @param callbacks Synthesis request callbacks. If null, the default request
		* callbacks object will be used.
		*/
		public void queueSpeak(final Markup markup,
		final UtteranceId utteranceId,
		final RequestConfig config,
		final RequestCallbacks callbacks) {
		runAction(new Action(ACTION_QUEUE_SPEAK_NAME) {
		@Override
		public void run(ITextToSpeechService service) throws RemoteException {
		@@ -908,7 +930,7 @@ public class TextToSpeechClient {

		int queueResult = service.speakV2(
		getCallerIdentity(),
		new SynthesisRequestV2(utterance, utteranceId.toUniqueString(), config));
		new SynthesisRequestV2(markup, utteranceId.toUniqueString(), config));
		if (queueResult != Status.SUCCESS) {
		removeCallbackAndErr(utteranceId.toUniqueString(), queueResult);
		}
		@@ -931,12 +953,37 @@ public class TextToSpeechClient {
		* @param outputFile File to write the generated audio data to.
		* @param config Synthesis request configuration. Can't be null. Have to contain a
		* voice.
		* @param callbacks Synthesis request callbacks. If null, default request
		* @param callbacks Synthesis request callbacks. If null, the default request
		* callbacks object will be used.
		*/
		public void queueSynthesizeToFile(final String utterance, final UtteranceId utteranceId,
		final File outputFile, final RequestConfig config,
		final RequestCallbacks callbacks) {
		queueSynthesizeToFile(createMarkupFromString(utterance), utteranceId, outputFile, config, callbacks);
		}

		/**
		* Synthesizes the given {@link Markup} (can be constructed with {@link Utterance})
		* to a file using the specified parameters. This method is asynchronous, i.e. the
		* method just adds the request to the queue of TTS requests and then returns. The
		* synthesis might not have finished (or even started!) at the time when this method
		* returns.
		*
		* @param markup The Markup that should be synthesized. The written equivalent of
		* the spoken text should be no longer than 1000 characters.
		* @param utteranceId Unique identificator used to track the synthesis progress
		* in {@link RequestCallbacks}.
		* @param outputFile File to write the generated audio data to.
		* @param config Synthesis request configuration. Can't be null. Have to contain a
		* voice.
		* @param callbacks Synthesis request callbacks. If null, the default request
		* callbacks object will be used.
		*/
		public void queueSynthesizeToFile(
		final Markup markup,
		final UtteranceId utteranceId,
		final File outputFile, final RequestConfig config,
		final RequestCallbacks callbacks) {
		runAction(new Action(ACTION_QUEUE_SYNTHESIZE_TO_FILE) {
		@Override
		public void run(ITextToSpeechService service) throws RemoteException {
		@@ -964,8 +1011,7 @@ public class TextToSpeechClient {

		int queueResult = service.synthesizeToFileDescriptorV2(getCallerIdentity(),
		fileDescriptor,
		new SynthesisRequestV2(utterance, utteranceId.toUniqueString(),
		config));
		new SynthesisRequestV2(markup, utteranceId.toUniqueString(), config));
		fileDescriptor.close();
		if (queueResult != Status.SUCCESS) {
		removeCallbackAndErr(utteranceId.toUniqueString(), queueResult);
		@@ -981,6 +1027,13 @@ public class TextToSpeechClient {
		});
		}

		private static Markup createMarkupFromString(String str) {
		return new Utterance()
		.append(new Utterance.TtsText(str))
		.setNoWarningOnFallback(true)
		.createMarkup();
		}

		private static final String ACTION_QUEUE_SILENCE_NAME = "queueSilence";

		/**

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

+49 −4

Original line number	Diff line number	Diff line
		@@ -352,6 +352,12 @@ public abstract class TextToSpeechService extends Service {
		params.putString(TextToSpeech.Engine.KEY_FEATURE_EMBEDDED_SYNTHESIS, "true");
		}

		String noWarning = request.getMarkup().getParameter(Utterance.KEY_NO_WARNING_ON_FALLBACK);
		if (noWarning == null \|\| noWarning.equals("false")) {
		Log.w("TextToSpeechService", "The synthesis engine does not support Markup, falling " +
		"back to the given plain text.");
		}

		// Build V1 request
		SynthesisRequest requestV1 = new SynthesisRequest(request.getText(), params);
		Locale locale = selectedVoice.getLocale();
		@@ -856,14 +862,53 @@ public abstract class TextToSpeechService extends Service {
		}
		}

		/**
		* Estimate of the character count equivalent of a Markup instance. Calculated
		* by summing the characters of all Markups of type "text". Each other node
		* is counted as a single character, as the character count of other nodes
		* is non-trivial to calculate and we don't want to accept arbitrarily large
		* requests.
		*/
		private int estimateSynthesisLengthFromMarkup(Markup m) {
		int size = 0;
		if (m.getType() != null &&
		m.getType().equals("text") &&
		m.getParameter("text") != null) {
		size += m.getParameter("text").length();
		} else if (m.getType() == null \|\|
		!m.getType().equals("utterance")) {
		size += 1;
		}
		for (Markup nested : m.getNestedMarkups()) {
		size += estimateSynthesisLengthFromMarkup(nested);
		}
		return size;
		}

		@Override
		public boolean isValid() {
		if (mSynthesisRequest.getText() == null) {
		Log.e(TAG, "null synthesis text");
		if (mSynthesisRequest.getMarkup() == null) {
		Log.e(TAG, "No markup in request.");
		return false;
		}
		if (mSynthesisRequest.getText().length() >= TextToSpeech.getMaxSpeechInputLength()) {
		Log.w(TAG, "Text too long: " + mSynthesisRequest.getText().length() + " chars");
		String type = mSynthesisRequest.getMarkup().getType();
		if (type == null) {
		Log.w(TAG, "Top level markup node should have type \"utterance\", not null");
		return false;
		} else if (!type.equals("utterance")) {
		Log.w(TAG, "Top level markup node should have type \"utterance\" instead of " +
		"\"" + type + "\"");
		return false;
		}

		int estimate = estimateSynthesisLengthFromMarkup(mSynthesisRequest.getMarkup());
		if (estimate >= TextToSpeech.getMaxSpeechInputLength()) {
		Log.w(TAG, "Text too long: estimated size of text was " + estimate + " chars.");
		return false;
		}

		if (estimate <= 0) {
		Log.e(TAG, "null synthesis text");
		return false;
		}