SoundTrigger API update. (d3b82233) · Commits · e / os / android_frameworks_base

core/java/android/hardware/soundtrigger/SoundTrigger.java

+246 −29

Original line number	Diff line number	Diff line
		@@ -16,6 +16,7 @@

		package android.hardware.soundtrigger;

		import android.media.AudioFormat;
		import android.os.Handler;
		import android.os.Parcel;
		import android.os.Parcelable;
		@@ -86,11 +87,15 @@ public class SoundTrigger {
		/** Rated power consumption when detection is active with TDB silence/sound/speech ratio */
		public final int powerConsumptionMw;

		/** Returns the trigger (key phrase) capture in the binary data of the
		* recognition callback event */
		public final boolean returnsTriggerInEvent;

		ModuleProperties(int id, String implementor, String description,
		String uuid, int version, int maxSoundModels, int maxKeyphrases,
		int maxUsers, int recognitionModes, boolean supportsCaptureTransition,
		int maxBufferMs, boolean supportsConcurrentCapture,
		int powerConsumptionMw) {
		int powerConsumptionMw, boolean returnsTriggerInEvent) {
		this.id = id;
		this.implementor = implementor;
		this.description = description;
		@@ -104,6 +109,7 @@ public class SoundTrigger {
		this.maxBufferMs = maxBufferMs;
		this.supportsConcurrentCapture = supportsConcurrentCapture;
		this.powerConsumptionMw = powerConsumptionMw;
		this.returnsTriggerInEvent = returnsTriggerInEvent;
		}

		public static final Parcelable.Creator<ModuleProperties> CREATOR
		@@ -131,10 +137,11 @@ public class SoundTrigger {
		int maxBufferMs = in.readInt();
		boolean supportsConcurrentCapture = in.readByte() == 1;
		int powerConsumptionMw = in.readInt();
		boolean returnsTriggerInEvent = in.readByte() == 1;
		return new ModuleProperties(id, implementor, description, uuid, version,
		maxSoundModels, maxKeyphrases, maxUsers, recognitionModes,
		supportsCaptureTransition, maxBufferMs, supportsConcurrentCapture,
		powerConsumptionMw);
		powerConsumptionMw, returnsTriggerInEvent);
		}

		@Override
		@@ -152,6 +159,7 @@ public class SoundTrigger {
		dest.writeInt(maxBufferMs);
		dest.writeByte((byte) (supportsConcurrentCapture ? 1 : 0));
		dest.writeInt(powerConsumptionMw);
		dest.writeByte((byte) (returnsTriggerInEvent ? 1 : 0));
		}

		@Override
		@@ -167,7 +175,8 @@ public class SoundTrigger {
		+ maxUsers + ", recognitionModes=" + recognitionModes
		+ ", supportsCaptureTransition=" + supportsCaptureTransition + ", maxBufferMs="
		+ maxBufferMs + ", supportsConcurrentCapture=" + supportsConcurrentCapture
		+ ", powerConsumptionMw=" + powerConsumptionMw + "]";
		+ ", powerConsumptionMw=" + powerConsumptionMw
		+ ", returnsTriggerInEvent=" + returnsTriggerInEvent + "]";
		}
		}

		@@ -190,11 +199,15 @@ public class SoundTrigger {
		/** Sound model type (e.g. TYPE_KEYPHRASE); */
		public final int type;

		/** Unique sound model vendor identifier */
		public final UUID vendorUuid;

		/** Opaque data. For use by vendor implementation and enrollment application */
		public final byte[] data;

		public SoundModel(UUID uuid, int type, byte[] data) {
		public SoundModel(UUID uuid, UUID vendorUuid, int type, byte[] data) {
		this.uuid = uuid;
		this.vendorUuid = vendorUuid;
		this.type = type;
		this.data = data;
		}
		@@ -329,8 +342,9 @@ public class SoundTrigger {
		/** Key phrases in this sound model */
		public final Keyphrase[] keyphrases; // keyword phrases in model

		public KeyphraseSoundModel(UUID id, byte[] data, Keyphrase[] keyphrases) {
		super(id, TYPE_KEYPHRASE, data);
		public KeyphraseSoundModel(
		UUID uuid, UUID vendorUuid, byte[] data, Keyphrase[] keyphrases) {
		super(uuid, vendorUuid, TYPE_KEYPHRASE, data);
		this.keyphrases = keyphrases;
		}

		@@ -347,9 +361,14 @@ public class SoundTrigger {

		private static KeyphraseSoundModel fromParcel(Parcel in) {
		UUID uuid = UUID.fromString(in.readString());
		UUID vendorUuid = null;
		int length = in.readInt();
		if (length >= 0) {
		vendorUuid = UUID.fromString(in.readString());
		}
		byte[] data = in.readBlob();
		Keyphrase[] keyphrases = in.createTypedArray(Keyphrase.CREATOR);
		return new KeyphraseSoundModel(uuid, data, keyphrases);
		return new KeyphraseSoundModel(uuid, vendorUuid, data, keyphrases);
		}

		@Override
		@@ -360,14 +379,21 @@ public class SoundTrigger {
		@Override
		public void writeToParcel(Parcel dest, int flags) {
		dest.writeString(uuid.toString());
		if (vendorUuid == null) {
		dest.writeInt(-1);
		} else {
		dest.writeInt(vendorUuid.toString().length());
		dest.writeString(vendorUuid.toString());
		}
		dest.writeBlob(data);
		dest.writeTypedArray(keyphrases, flags);
		}

		@Override
		public String toString() {
		return "KeyphraseSoundModel [keyphrases=" + Arrays.toString(keyphrases) + ", uuid="
		+ uuid + ", type=" + type + ", data=" + (data == null ? 0 : data.length) + "]";
		return "KeyphraseSoundModel [keyphrases=" + Arrays.toString(keyphrases)
		+ ", uuid=" + uuid + ", vendorUuid=" + vendorUuid
		+ ", type=" + type + ", data=" + (data == null ? 0 : data.length) + "]";
		}
		}

		@@ -411,18 +437,26 @@ public class SoundTrigger {
		public final int captureDelayMs;
		/** Duration in ms of audio captured before the start of the trigger. 0 if none. */
		public final int capturePreambleMs;
		/** True if the trigger (key phrase capture is present in binary data */
		public final boolean triggerInData;
		/** Audio format of either the trigger in event data or to use for capture of the
		* rest of the utterance */
		public AudioFormat captureFormat;
		/** Opaque data for use by system applications who know about voice engine internals,
		* typically during enrollment. */
		public final byte[] data;

		public RecognitionEvent(int status, int soundModelHandle, boolean captureAvailable,
		int captureSession, int captureDelayMs, int capturePreambleMs, byte[] data) {
		int captureSession, int captureDelayMs, int capturePreambleMs,
		boolean triggerInData, AudioFormat captureFormat, byte[] data) {
		this.status = status;
		this.soundModelHandle = soundModelHandle;
		this.captureAvailable = captureAvailable;
		this.captureSession = captureSession;
		this.captureDelayMs = captureDelayMs;
		this.capturePreambleMs = capturePreambleMs;
		this.triggerInData = triggerInData;
		this.captureFormat = captureFormat;
		this.data = data;
		}

		@@ -444,9 +478,21 @@ public class SoundTrigger {
		int captureSession = in.readInt();
		int captureDelayMs = in.readInt();
		int capturePreambleMs = in.readInt();
		boolean triggerInData = in.readByte() == 1;
		AudioFormat captureFormat = null;
		if (triggerInData) {
		int sampleRate = in.readInt();
		int encoding = in.readInt();
		int channelMask = in.readInt();
		captureFormat = (new AudioFormat.Builder())
		.setChannelMask(channelMask)
		.setEncoding(encoding)
		.setSampleRate(sampleRate)
		.build();
		}
		byte[] data = in.readBlob();
		return new RecognitionEvent(status, soundModelHandle, captureAvailable, captureSession,
		captureDelayMs, capturePreambleMs, data);
		captureDelayMs, capturePreambleMs, triggerInData, captureFormat, data);
		}

		@Override
		@@ -462,6 +508,14 @@ public class SoundTrigger {
		dest.writeInt(captureSession);
		dest.writeInt(captureDelayMs);
		dest.writeInt(capturePreambleMs);
		if (triggerInData && (captureFormat != null)) {
		dest.writeByte((byte)1);
		dest.writeInt(captureFormat.getSampleRate());
		dest.writeInt(captureFormat.getEncoding());
		dest.writeInt(captureFormat.getChannelMask());
		} else {
		dest.writeByte((byte)0);
		}
		dest.writeBlob(data);
		}

		@@ -473,6 +527,12 @@ public class SoundTrigger {
		result = prime * result + captureDelayMs;
		result = prime * result + capturePreambleMs;
		result = prime * result + captureSession;
		result = prime * result + (triggerInData ? 1231 : 1237);
		if (captureFormat != null) {
		result = prime * result + captureFormat.getSampleRate();
		result = prime * result + captureFormat.getEncoding();
		result = prime * result + captureFormat.getChannelMask();
		}
		result = prime * result + Arrays.hashCode(data);
		result = prime * result + soundModelHandle;
		result = prime * result + status;
		@@ -502,6 +562,14 @@ public class SoundTrigger {
		return false;
		if (status != other.status)
		return false;
		if (triggerInData != other.triggerInData)
		return false;
		if (captureFormat.getSampleRate() != other.captureFormat.getSampleRate())
		return false;
		if (captureFormat.getEncoding() != other.captureFormat.getEncoding())
		return false;
		if (captureFormat.getChannelMask() != other.captureFormat.getChannelMask())
		return false;
		return true;
		}

		@@ -511,6 +579,13 @@ public class SoundTrigger {
		+ ", captureAvailable=" + captureAvailable + ", captureSession="
		+ captureSession + ", captureDelayMs=" + captureDelayMs
		+ ", capturePreambleMs=" + capturePreambleMs
		+ ", triggerInData=" + triggerInData
		+ ((captureFormat == null) ? "" :
		(", sampleRate=" + captureFormat.getSampleRate()))
		+ ((captureFormat == null) ? "" :
		(", encoding=" + captureFormat.getEncoding()))
		+ ((captureFormat == null) ? "" :
		(", channelMask=" + captureFormat.getChannelMask()))
		+ ", data=" + (data == null ? 0 : data.length) + "]";
		}
		}
		@@ -673,14 +748,19 @@ public class SoundTrigger {
		/** Recognition modes matched for this event */
		public final int recognitionModes;

		/** Confidence level for mode RECOGNITION_MODE_VOICE_TRIGGER when user identification
		* is not performed */
		public final int coarseConfidenceLevel;

		/** Confidence levels for all users recognized (KeyphraseRecognitionEvent) or to
		* be recognized (RecognitionConfig) */
		public final ConfidenceLevel[] confidenceLevels;

		public KeyphraseRecognitionExtra(int id, int recognitionModes,
		public KeyphraseRecognitionExtra(int id, int recognitionModes, int coarseConfidenceLevel,
		ConfidenceLevel[] confidenceLevels) {
		this.id = id;
		this.recognitionModes = recognitionModes;
		this.coarseConfidenceLevel = coarseConfidenceLevel;
		this.confidenceLevels = confidenceLevels;
		}

		@@ -698,14 +778,17 @@ public class SoundTrigger {
		private static KeyphraseRecognitionExtra fromParcel(Parcel in) {
		int id = in.readInt();
		int recognitionModes = in.readInt();
		int coarseConfidenceLevel = in.readInt();
		ConfidenceLevel[] confidenceLevels = in.createTypedArray(ConfidenceLevel.CREATOR);
		return new KeyphraseRecognitionExtra(id, recognitionModes, confidenceLevels);
		return new KeyphraseRecognitionExtra(id, recognitionModes, coarseConfidenceLevel,
		confidenceLevels);
		}

		@Override
		public void writeToParcel(Parcel dest, int flags) {
		dest.writeInt(id);
		dest.writeInt(recognitionModes);
		dest.writeInt(coarseConfidenceLevel);
		dest.writeTypedArray(confidenceLevels, flags);
		}

		@@ -721,6 +804,7 @@ public class SoundTrigger {
		result = prime * result + Arrays.hashCode(confidenceLevels);
		result = prime * result + id;
		result = prime * result + recognitionModes;
		result = prime * result + coarseConfidenceLevel;
		return result;
		}

		@@ -739,12 +823,15 @@ public class SoundTrigger {
		return false;
		if (recognitionModes != other.recognitionModes)
		return false;
		if (coarseConfidenceLevel != other.coarseConfidenceLevel)
		return false;
		return true;
		}

		@Override
		public String toString() {
		return "KeyphraseRecognitionExtra [id=" + id + ", recognitionModes=" + recognitionModes
		+ ", coarseConfidenceLevel=" + coarseConfidenceLevel
		+ ", confidenceLevels=" + Arrays.toString(confidenceLevels) + "]";
		}
		}
		@@ -756,15 +843,12 @@ public class SoundTrigger {
		/** Indicates if the key phrase is present in the buffered audio available for capture */
		public final KeyphraseRecognitionExtra[] keyphraseExtras;

		/** Additional data available for each recognized key phrases in the model */
		public final boolean keyphraseInCapture;

		public KeyphraseRecognitionEvent(int status, int soundModelHandle, boolean captureAvailable,
		int captureSession, int captureDelayMs, int capturePreambleMs, byte[] data,
		boolean keyphraseInCapture, KeyphraseRecognitionExtra[] keyphraseExtras) {
		int captureSession, int captureDelayMs, int capturePreambleMs,
		boolean triggerInData, AudioFormat captureFormat, byte[] data,
		KeyphraseRecognitionExtra[] keyphraseExtras) {
		super(status, soundModelHandle, captureAvailable, captureSession, captureDelayMs,
		capturePreambleMs, data);
		this.keyphraseInCapture = keyphraseInCapture;
		capturePreambleMs, triggerInData, captureFormat, data);
		this.keyphraseExtras = keyphraseExtras;
		}

		@@ -786,13 +870,24 @@ public class SoundTrigger {
		int captureSession = in.readInt();
		int captureDelayMs = in.readInt();
		int capturePreambleMs = in.readInt();
		boolean triggerInData = in.readByte() == 1;
		AudioFormat captureFormat = null;
		if (triggerInData) {
		int sampleRate = in.readInt();
		int encoding = in.readInt();
		int channelMask = in.readInt();
		captureFormat = (new AudioFormat.Builder())
		.setChannelMask(channelMask)
		.setEncoding(encoding)
		.setSampleRate(sampleRate)
		.build();
		}
		byte[] data = in.readBlob();
		boolean keyphraseInCapture = in.readByte() == 1;
		KeyphraseRecognitionExtra[] keyphraseExtras =
		in.createTypedArray(KeyphraseRecognitionExtra.CREATOR);
		return new KeyphraseRecognitionEvent(status, soundModelHandle, captureAvailable,
		captureSession, captureDelayMs, capturePreambleMs, data, keyphraseInCapture,
		keyphraseExtras);
		captureSession, captureDelayMs, capturePreambleMs, triggerInData,
		captureFormat, data, keyphraseExtras);
		}

		@Override
		@@ -803,8 +898,15 @@ public class SoundTrigger {
		dest.writeInt(captureSession);
		dest.writeInt(captureDelayMs);
		dest.writeInt(capturePreambleMs);
		if (triggerInData && (captureFormat != null)) {
		dest.writeByte((byte)1);
		dest.writeInt(captureFormat.getSampleRate());
		dest.writeInt(captureFormat.getEncoding());
		dest.writeInt(captureFormat.getChannelMask());
		} else {
		dest.writeByte((byte)0);
		}
		dest.writeBlob(data);
		dest.writeByte((byte) (keyphraseInCapture ? 1 : 0));
		dest.writeTypedArray(keyphraseExtras, flags);
		}

		@@ -818,7 +920,6 @@ public class SoundTrigger {
		final int prime = 31;
		int result = super.hashCode();
		result = prime * result + Arrays.hashCode(keyphraseExtras);
		result = prime * result + (keyphraseInCapture ? 1231 : 1237);
		return result;
		}

		@@ -833,22 +934,126 @@ public class SoundTrigger {
		KeyphraseRecognitionEvent other = (KeyphraseRecognitionEvent) obj;
		if (!Arrays.equals(keyphraseExtras, other.keyphraseExtras))
		return false;
		if (keyphraseInCapture != other.keyphraseInCapture)
		return false;
		return true;
		}

		@Override
		public String toString() {
		return "KeyphraseRecognitionEvent [keyphraseExtras=" + Arrays.toString(keyphraseExtras)
		+ ", keyphraseInCapture=" + keyphraseInCapture + ", status=" + status
		+ ", status=" + status
		+ ", soundModelHandle=" + soundModelHandle + ", captureAvailable="
		+ captureAvailable + ", captureSession=" + captureSession + ", captureDelayMs="
		+ captureDelayMs + ", capturePreambleMs=" + capturePreambleMs
		+ ", triggerInData=" + triggerInData
		+ ((captureFormat == null) ? "" :
		(", sampleRate=" + captureFormat.getSampleRate()))
		+ ((captureFormat == null) ? "" :
		(", encoding=" + captureFormat.getEncoding()))
		+ ((captureFormat == null) ? "" :
		(", channelMask=" + captureFormat.getChannelMask()))
		+ ", data=" + (data == null ? 0 : data.length) + "]";
		}
		}

		/**
		* Status codes for {@link SoundModelEvent}
		*/
		/** Sound Model was updated */
		public static final int SOUNDMODEL_STATUS_UPDATED = 0;

		/**
		* A SoundModelEvent is provided by the
		* {@link StatusListener#onSoundModelUpdate(SoundModelEvent)}
		* callback when a sound model has been updated by the implementation
		*/
		public static class SoundModelEvent implements Parcelable {
		/** Status e.g {@link #SOUNDMODEL_STATUS_UPDATED} */
		public final int status;
		/** The updated sound model handle */
		public final int soundModelHandle;
		/** New sound model data */
		public final byte[] data;

		SoundModelEvent(int status, int soundModelHandle, byte[] data) {
		this.status = status;
		this.soundModelHandle = soundModelHandle;
		this.data = data;
		}

		public static final Parcelable.Creator<SoundModelEvent> CREATOR
		= new Parcelable.Creator<SoundModelEvent>() {
		public SoundModelEvent createFromParcel(Parcel in) {
		return SoundModelEvent.fromParcel(in);
		}

		public SoundModelEvent[] newArray(int size) {
		return new SoundModelEvent[size];
		}
		};

		private static SoundModelEvent fromParcel(Parcel in) {
		int status = in.readInt();
		int soundModelHandle = in.readInt();
		byte[] data = in.readBlob();
		return new SoundModelEvent(status, soundModelHandle, data);
		}

		@Override
		public int describeContents() {
		return 0;
		}

		@Override
		public void writeToParcel(Parcel dest, int flags) {
		dest.writeInt(status);
		dest.writeInt(soundModelHandle);
		dest.writeBlob(data);
		}

		@Override
		public int hashCode() {
		final int prime = 31;
		int result = 1;
		result = prime * result + Arrays.hashCode(data);
		result = prime * result + soundModelHandle;
		result = prime * result + status;
		return result;
		}

		@Override
		public boolean equals(Object obj) {
		if (this == obj)
		return true;
		if (obj == null)
		return false;
		if (getClass() != obj.getClass())
		return false;
		SoundModelEvent other = (SoundModelEvent) obj;
		if (!Arrays.equals(data, other.data))
		return false;
		if (soundModelHandle != other.soundModelHandle)
		return false;
		if (status != other.status)
		return false;
		return true;
		}

		@Override
		public String toString() {
		return "SoundModelEvent [status=" + status + ", soundModelHandle=" + soundModelHandle
		+ ", data=" + (data == null ? 0 : data.length) + "]";
		}
		}

		/**
		* Native service state. {@link StatusListener#onServiceStateChange(int)}
		*/
		// Keep in sync with system/core/include/system/sound_trigger.h
		/** Sound trigger service is enabled */
		public static final int SERVICE_STATE_ENABLED = 0;
		/** Sound trigger service is disabled */
		public static final int SERVICE_STATE_DISABLED = 1;

		/**
		* Returns a list of descriptors for all harware modules loaded.
		* @param modules A ModuleProperties array where the list will be returned.
		@@ -890,6 +1095,18 @@ public class SoundTrigger {
		*/
		public abstract void onRecognition(RecognitionEvent event);

		/**
		* Called when a sound model has been updated
		*/
		public abstract void onSoundModelUpdate(SoundModelEvent event);

		/**
		* Called when the sound trigger native service state changes.
		* @param state Native service state. One of {@link SoundTrigger#SERVICE_STATE_ENABLED},
		* {@link SoundTrigger#SERVICE_STATE_DISABLED}
		*/
		public abstract void onServiceStateChange(int state);

		/**
		* Called when the sound trigger native service dies
		*/

core/java/android/hardware/soundtrigger/SoundTriggerModule.java

+15 −4

Original line number	Diff line number	Diff line
		@@ -36,8 +36,11 @@ public class SoundTriggerModule {
		private int mId;
		private NativeEventHandlerDelegate mEventHandlerDelegate;

		// to be kept in sync with core/jni/android_hardware_SoundTrigger.cpp
		private static final int EVENT_RECOGNITION = 1;
		private static final int EVENT_SERVICE_DIED = 2;
		private static final int EVENT_SOUNDMODEL = 3;
		private static final int EVENT_SERVICE_STATE_CHANGE = 4;

		SoundTriggerModule(int moduleId, SoundTrigger.StatusListener listener, Handler handler) {
		mId = moduleId;
		@@ -133,11 +136,8 @@ public class SoundTriggerModule {
		if (handler != null) {
		looper = handler.getLooper();
		} else {
		looper = Looper.myLooper();
		if (looper == null) {
		looper = Looper.getMainLooper();
		}
		}

		// construct the event handler with this looper
		if (looper != null) {
		@@ -152,6 +152,17 @@ public class SoundTriggerModule {
		(SoundTrigger.RecognitionEvent)msg.obj);
		}
		break;
		case EVENT_SOUNDMODEL:
		if (listener != null) {
		listener.onSoundModelUpdate(
		(SoundTrigger.SoundModelEvent)msg.obj);
		}
		break;
		case EVENT_SERVICE_STATE_CHANGE:
		if (listener != null) {
		listener.onServiceStateChange(msg.arg1);
		}
		break;
		case EVENT_SERVICE_DIED:
		if (listener != null) {
		listener.onServiceDied();

core/java/android/service/voice/AlwaysOnHotwordDetector.java

+1 −1

Original line number	Diff line number	Diff line
		@@ -426,7 +426,7 @@ public class AlwaysOnHotwordDetector {
		KeyphraseRecognitionExtra[] recognitionExtra = new KeyphraseRecognitionExtra[1];
		// TODO: Do we need to do something about the confidence level here?
		recognitionExtra[0] = new KeyphraseRecognitionExtra(mKeyphraseMetadata.id,
		mKeyphraseMetadata.recognitionModeFlags, new ConfidenceLevel[0]);
		mKeyphraseMetadata.recognitionModeFlags, 0, new ConfidenceLevel[0]);
		boolean captureTriggerAudio =
		(recognitionFlags&RECOGNITION_FLAG_CAPTURE_TRIGGER_AUDIO) != 0;
		boolean allowMultipleTriggers =

core/jni/android_hardware_SoundTrigger.cpp

+118 −18

File changed.

Preview size limit exceeded, changes collapsed.

media/java/android/media/AudioFormat.java

+14 −0

Original line number	Diff line number	Diff line
		@@ -249,6 +249,20 @@ public class AudioFormat {
		private AudioFormat(int ignoredArgument) {
		}

		/**
		* Constructor used by the JNI
		*/
		// Update sound trigger JNI in core/jni/android_hardware_SoundTrigger.cpp when modifying this
		// constructor
		private AudioFormat(int encoding, int sampleRate, int channelMask) {
		mEncoding = encoding;
		mSampleRate = sampleRate;
		mChannelMask = channelMask;
		mPropertySetMask = AUDIO_FORMAT_HAS_PROPERTY_ENCODING \|
		AUDIO_FORMAT_HAS_PROPERTY_SAMPLE_RATE \|
		AUDIO_FORMAT_HAS_PROPERTY_CHANNEL_MASK;
		}

		/** @hide */
		public final static int AUDIO_FORMAT_HAS_PROPERTY_NONE = 0x0;
		/** @hide */