net.idrnd.voicesdk.diarization

Class DiarizationEngine

java.lang.Object

net.idrnd.voicesdk.common.VoiceSdkNativePeer

net.idrnd.voicesdk.diarization.DiarizationEngine

All Implemented Interfaces:

java.lang.AutoCloseable

public class DiarizationEngine
extends VoiceSdkNativePeer

Performs the diarization split of the arbitrary speech data. Diarization means dividing speech audio data into speakers and the intervals when they spoke in the input data.

This class serves as gateway to native Voice SDK implementation and allocates resources on native heap. To release the allocated memory, AutoCloseable.close() method should be invoked when the instance is no longer needed.

Any method that delegates to native call may throw VoiceSdkEngineException

Field Summary

Fields inherited from class net.idrnd.voicesdk.common.VoiceSdkNativePeer

nativeId

Constructor Summary

Constructors

Constructor and Description
DiarizationEngine(java.lang.String initDataPath) Sole constructor.

Method Summary

Modifier and Type	Method and Description
DiarizationResult	getSegmentation(byte[] pcm16Bytes, int sampleRate) Performs diarization on audio bytes provided in PCM16 format.
DiarizationResult	getSegmentation(byte[] pcm16Bytes, int sampleRate, int numSpeakers) Performs diarization on audio bytes provided in PCM16 format.
DiarizationResult	getSegmentation(float[] floatSamples, int sampleRate) Performs diarization on audio provided in normalized float format.
DiarizationResult	getSegmentation(float[] floatSamples, int sampleRate, int numSpeakers) Performs diarization on audio provided in normalized float format.
DiarizationResult	getSegmentation(short[] pcm16Samples, int sampleRate) Performs diarization on audio provided in PCM16 format.
DiarizationResult	getSegmentation(short[] pcm16Samples, int sampleRate, int numSpeakers) Performs diarization on audio provided in PCM16 format.
DiarizationResult	getSegmentation(java.lang.String audioFile) Performs diarization on audio file.
DiarizationResult	getSegmentation(java.lang.String audioFile, int numSpeakers) Performs diarization on audio file.
protected DiarizationResult	getSegmentation1(byte[] pcm16Bytes, int sampleRate, int numSpeakers)
protected DiarizationResult	getSegmentation2(short[] pcm16Samples, int sampleRate, int numSpeakers)
protected DiarizationResult	getSegmentation3(float[] floatSamples, int sampleRate, int numSpeakers)
protected DiarizationResult	getSegmentation4(java.lang.String audioFile, int numSpeakers)
protected static long	init(java.lang.String initDataPath)
protected void	release() This method should release resources on native layer (indirectly using nativeId)

Methods inherited from class net.idrnd.voicesdk.common.VoiceSdkNativePeer

close, equals, finalize, hashCode

Methods inherited from class java.lang.Object

clone, getClass, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

DiarizationEngine

public DiarizationEngine(java.lang.String initDataPath)

Sole constructor. Initializes the Engine object and allocates the native resources.

Parameters:

initDataPath - full path to the directory with SDK data

Throws:

VoiceSdkEngineException - wraps native exceptions

Method Detail

getSegmentation

public DiarizationResult getSegmentation(java.lang.String audioFile)

Performs diarization on audio file. Number of speakers is arbitrary.

Parameters:

audioFile - Path to the audio file

Returns:

DiarizationResult

Throws:

VoiceSdkEngineException - wraps native exceptions

getSegmentation

public DiarizationResult getSegmentation(java.lang.String audioFile,
                                         int numSpeakers)

Performs diarization on audio file. Number of speakers is fixed.

Parameters:

audioFile - Path to the audio file

numSpeakers - max number of speakers to detect

Returns:

DiarizationResult

Throws:

VoiceSdkEngineException - wraps native exceptions

getSegmentation

public DiarizationResult getSegmentation(float[] floatSamples,
                                         int sampleRate)

Performs diarization on audio provided in normalized float format. Number of speakers is arbitrary.

Parameters:

floatSamples - Array of float audio samples (in [-1, 1] range)

sampleRate - Sample rate of the audio, Hz

Returns:

DiarizationResult

Throws:

VoiceSdkEngineException - wraps native exceptions

getSegmentation

public DiarizationResult getSegmentation(float[] floatSamples,
                                         int sampleRate,
                                         int numSpeakers)

Performs diarization on audio provided in normalized float format. Number of speakers is fixed.

Parameters:

floatSamples - Array of float audio samples (in [-1, 1] range)

sampleRate - Sample rate of the audio, Hz

numSpeakers - max number of speakers to detect

Returns:

DiarizationResult

Throws:

VoiceSdkEngineException - wraps native exceptions

getSegmentation

public DiarizationResult getSegmentation(short[] pcm16Samples,
                                         int sampleRate)

Performs diarization on audio provided in PCM16 format. Number of speakers is arbitrary.

Parameters:

pcm16Samples - Array of PCM16 audio samples

sampleRate - Sample rate of the audio, Hz

Returns:

DiarizationResult

Throws:

VoiceSdkEngineException - wraps native exceptions

getSegmentation

public DiarizationResult getSegmentation(short[] pcm16Samples,
                                         int sampleRate,
                                         int numSpeakers)

Performs diarization on audio provided in PCM16 format. Number of speakers is fixed.

Parameters:

pcm16Samples - Array of PCM16 audio samples

sampleRate - Sample rate of the audio, Hz

numSpeakers - max number of speakers to detect

Returns:

DiarizationResult

Throws:

VoiceSdkEngineException - wraps native exceptions

getSegmentation

public DiarizationResult getSegmentation(byte[] pcm16Bytes,
                                         int sampleRate)

Performs diarization on audio bytes provided in PCM16 format. Number of speakers is arbitrary.

Parameters:

pcm16Bytes - Array of little-endian PCM16 audio bytes

sampleRate - Sample rate of the audio, Hz

Returns:

DiarizationResult

Throws:

VoiceSdkEngineException - wraps native exceptions

getSegmentation

public DiarizationResult getSegmentation(byte[] pcm16Bytes,
                                         int sampleRate,
                                         int numSpeakers)

Performs diarization on audio bytes provided in PCM16 format. Number of speakers is fixed.

Parameters:

pcm16Bytes - Array of little-endian PCM16 audio bytes

sampleRate - Sample rate of the audio, Hz

numSpeakers - max number of speakers to detect

Returns:

DiarizationResult

Throws:

VoiceSdkEngineException - wraps native exceptions

init

protected static long init(java.lang.String initDataPath)

getSegmentation4

protected DiarizationResult getSegmentation4(java.lang.String audioFile,
                                             int numSpeakers)

getSegmentation3

protected DiarizationResult getSegmentation3(float[] floatSamples,
                                             int sampleRate,
                                             int numSpeakers)

getSegmentation2

protected DiarizationResult getSegmentation2(short[] pcm16Samples,
                                             int sampleRate,
                                             int numSpeakers)

getSegmentation1

protected DiarizationResult getSegmentation1(byte[] pcm16Bytes,
                                             int sampleRate,
                                             int numSpeakers)

release

protected void release()

Description copied from class: VoiceSdkNativePeer

This method should release resources on native layer (indirectly using nativeId)

Specified by:

release in class VoiceSdkNativePeer