io.agora.rtc

Class AgoraAudioVadConfig

java.lang.Object

io.agora.rtc.AgoraAudioVadConfig

public class AgoraAudioVadConfig
extends java.lang.Object

Configuration values for the Java VAD implementation.

Field Summary

Fields

Modifier and Type	Field and Description
private float	activePercent active percent, if over this percent, will be recognized as speaking, default value is 0.6
private float	aggressive aggressive factor, greater value means more aggressive, default value is 5.0
private int	anaWindowSz fft-window Size, will be used to calc rms, default value is 768
private int	fftSz fft-size, only support: 128, 256, 512, 1024, default value is 1024
private int	frqInputAvailableFlag whether Aed_InputData will contain external freq.
private int	hopSz fft-Hop Size, will be used to check, default value is 160
private float	inactivePercent inactive percent, if below this percent, will be recognized as non-speaking, default value is 0.2
private float	jointThr joint threshold in dB, default value is 0.0
private int	preStartRecognizeCount pre start recognize count, buffer size for 10ms 16KHz 16bit 1channel PCM, default value is 10
private float	rmsThr rms threshold in dB, default value is -40.0
private int	startRecognizeCount start recognize count, buffer size for 10ms 16KHz 16bit 1channel PCM, default value is 10
private int	stopRecognizeCount stop recognize count, buffer size for 10ms 16KHz 16bit 1channel PCM, default value is 6
private int	useCVersionAIModule whether to use the C version of AI submodules, default value is 0
private float	voiceProbThr voice probability threshold 0.0f ~ 1.0f, default value is 0.8

Constructor Summary

Constructors

Constructor and Description
AgoraAudioVadConfig() Creates a configuration with default thresholds and counters.
AgoraAudioVadConfig(int fftSz, int hopSz, int anaWindowSz, int frqInputAvailableFlag, int useCVersionAIModule, float voiceProbThr, float rmsThr, float jointThr, float aggressive, int startRecognizeCount, int stopRecognizeCount, int preStartRecognizeCount, float activePercent, float inactivePercent) Creates a configuration with custom parameters.

Method Summary

Modifier and Type	Method and Description
float	getActivePercent() Returns the percentage threshold for active detection.
float	getAggressive() Returns the aggressiveness factor.
int	getAnaWindowSz() Returns the analysis window size.
int	getFftSz() Returns the FFT size used by the detector.
int	getFrqInputAvailableFlag() Returns the flag indicating external frequency input.
int	getHopSz() Returns the FFT hop size.
float	getInactivePercent() Returns the percentage threshold for inactivity detection.
float	getJointThr() Returns the joint threshold in dB.
int	getPreStartRecognizeCount() Returns the pre-start recognition buffer size.
float	getRmsThr() Returns the RMS threshold in dB.
int	getStartRecognizeCount() Returns the frames required to start recognition.
int	getStopRecognizeCount() Returns the frames required to stop recognition.
int	getUseCVersionAIModule() Returns the flag for using the C AI modules.
float	getVoiceProbThr() Returns the voice probability threshold.
void	setActivePercent(float activePercent) Sets the percentage threshold for active detection.
void	setAggressive(float aggressive) Sets the aggressiveness factor.
void	setAnaWindowSz(int anaWindowSz) Sets the analysis window size.
void	setFftSz(int fftSz) Sets the FFT size used by the detector.
void	setFrqInputAvailableFlag(int frqInputAvailableFlag) Sets the flag indicating external frequency input.
void	setHopSz(int hopSz) Sets the FFT hop size.
void	setInactivePercent(float inactivePercent) Sets the percentage threshold for inactivity detection.
void	setJointThr(float jointThr) Sets the joint threshold in dB.
void	setPreStartRecognizeCount(int preStartRecognizeCount) Sets the pre-start recognition buffer size.
void	setRmsThr(float rmsThr) Sets the RMS threshold in dB.
void	setStartRecognizeCount(int startRecognizeCount) Sets the frames required to start recognition.
void	setStopRecognizeCount(int stopRecognizeCount) Sets the frames required to stop recognition.
void	setUseCVersionAIModule(int useCVersionAIModule) Sets the flag for using the C AI modules.
void	setVoiceProbThr(float voiceProbThr) Sets the voice probability threshold.
java.lang.String	toString()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

fftSz

private int fftSz

fft-size, only support: 128, 256, 512, 1024, default value is 1024

hopSz

private int hopSz

fft-Hop Size, will be used to check, default value is 160

anaWindowSz

private int anaWindowSz

fft-window Size, will be used to calc rms, default value is 768

frqInputAvailableFlag

private int frqInputAvailableFlag

whether Aed_InputData will contain external freq. power-sepctra, default value is 0

useCVersionAIModule

private int useCVersionAIModule

whether to use the C version of AI submodules, default value is 0

voiceProbThr

private float voiceProbThr

voice probability threshold 0.0f ~ 1.0f, default value is 0.8

rmsThr

private float rmsThr

rms threshold in dB, default value is -40.0

jointThr

private float jointThr

joint threshold in dB, default value is 0.0

aggressive

private float aggressive

aggressive factor, greater value means more aggressive, default value is 5.0

startRecognizeCount

private int startRecognizeCount

start recognize count, buffer size for 10ms 16KHz 16bit 1channel PCM, default value is 10

stopRecognizeCount

private int stopRecognizeCount

stop recognize count, buffer size for 10ms 16KHz 16bit 1channel PCM, default value is 6

preStartRecognizeCount

private int preStartRecognizeCount

pre start recognize count, buffer size for 10ms 16KHz 16bit 1channel PCM, default value is 10

activePercent

private float activePercent

active percent, if over this percent, will be recognized as speaking, default value is 0.6

inactivePercent

private float inactivePercent

inactive percent, if below this percent, will be recognized as non-speaking, default value is 0.2

Constructor Detail

AgoraAudioVadConfig

public AgoraAudioVadConfig()

Creates a configuration with default thresholds and counters.

AgoraAudioVadConfig

public AgoraAudioVadConfig(int fftSz,
                           int hopSz,
                           int anaWindowSz,
                           int frqInputAvailableFlag,
                           int useCVersionAIModule,
                           float voiceProbThr,
                           float rmsThr,
                           float jointThr,
                           float aggressive,
                           int startRecognizeCount,
                           int stopRecognizeCount,
                           int preStartRecognizeCount,
                           float activePercent,
                           float inactivePercent)

Creates a configuration with custom parameters.

Parameters:

fftSz - FFT window size

hopSz - Hop size for FFT

anaWindowSz - Window size used to calculate RMS

frqInputAvailableFlag - Flag indicating external frequency spectra availability

useCVersionAIModule - Flag indicating whether to use C AI modules

voiceProbThr - Voice probability threshold

rmsThr - RMS threshold in dB

jointThr - Joint threshold in dB

aggressive - Aggressiveness factor

startRecognizeCount - Frames required to enter speaking state

stopRecognizeCount - Frames required to leave speaking state

preStartRecognizeCount - Frames buffered before recognizing speaking

activePercent - Percentage threshold to treat as speaking

inactivePercent - Percentage threshold to treat as silence

Method Detail

getFftSz

public int getFftSz()

Returns the FFT size used by the detector.

Returns:

FFT size

setFftSz

public void setFftSz(int fftSz)

Sets the FFT size used by the detector.

Parameters:

fftSz - FFT size

getHopSz

public int getHopSz()

Returns the FFT hop size.

Returns:

Hop size

setHopSz

public void setHopSz(int hopSz)

Sets the FFT hop size.

Parameters:

hopSz - Hop size

getAnaWindowSz

public int getAnaWindowSz()

Returns the analysis window size.

Returns:

Analysis window size

setAnaWindowSz

public void setAnaWindowSz(int anaWindowSz)

Sets the analysis window size.

Parameters:

anaWindowSz - Analysis window size

getFrqInputAvailableFlag

public int getFrqInputAvailableFlag()

Returns the flag indicating external frequency input.

Returns:

Frequency input flag

setFrqInputAvailableFlag

public void setFrqInputAvailableFlag(int frqInputAvailableFlag)

Sets the flag indicating external frequency input.

Parameters:

frqInputAvailableFlag - Frequency input flag

getUseCVersionAIModule

public int getUseCVersionAIModule()

Returns the flag for using the C AI modules.

Returns:

C AI module flag

setUseCVersionAIModule

public void setUseCVersionAIModule(int useCVersionAIModule)

Sets the flag for using the C AI modules.

Parameters:

useCVersionAIModule - C AI module flag

getVoiceProbThr

public float getVoiceProbThr()

Returns the voice probability threshold.

Returns:

Voice probability threshold

setVoiceProbThr

public void setVoiceProbThr(float voiceProbThr)

Sets the voice probability threshold.

Parameters:

voiceProbThr - Voice probability threshold

getRmsThr

public float getRmsThr()

Returns the RMS threshold in dB.

Returns:

RMS threshold

setRmsThr

public void setRmsThr(float rmsThr)

Sets the RMS threshold in dB.

Parameters:

rmsThr - RMS threshold

getJointThr

public float getJointThr()

Returns the joint threshold in dB.

Returns:

Joint threshold

setJointThr

public void setJointThr(float jointThr)

Sets the joint threshold in dB.

Parameters:

jointThr - Joint threshold

getAggressive

public float getAggressive()

Returns the aggressiveness factor.

Returns:

Aggressiveness factor

setAggressive

public void setAggressive(float aggressive)

Sets the aggressiveness factor.

Parameters:

aggressive - Aggressiveness factor

getStartRecognizeCount

public int getStartRecognizeCount()

Returns the frames required to start recognition.

Returns:

Start recognition count

setStartRecognizeCount

public void setStartRecognizeCount(int startRecognizeCount)

Sets the frames required to start recognition.

Parameters:

startRecognizeCount - Start recognition count

getStopRecognizeCount

public int getStopRecognizeCount()

Returns the frames required to stop recognition.

Returns:

Stop recognition count

setStopRecognizeCount

public void setStopRecognizeCount(int stopRecognizeCount)

Sets the frames required to stop recognition.

Parameters:

stopRecognizeCount - Stop recognition count

getPreStartRecognizeCount

public int getPreStartRecognizeCount()

Returns the pre-start recognition buffer size.

Returns:

Pre-start recognition count

setPreStartRecognizeCount

public void setPreStartRecognizeCount(int preStartRecognizeCount)

Sets the pre-start recognition buffer size.

Parameters:

preStartRecognizeCount - Pre-start recognition count

getActivePercent

public float getActivePercent()

Returns the percentage threshold for active detection.

Returns:

Active percentage

setActivePercent

public void setActivePercent(float activePercent)

Sets the percentage threshold for active detection.

Parameters:

activePercent - Active percentage

getInactivePercent

public float getInactivePercent()

Returns the percentage threshold for inactivity detection.

Returns:

Inactive percentage

setInactivePercent

public void setInactivePercent(float inactivePercent)

Sets the percentage threshold for inactivity detection.

Parameters:

inactivePercent - Inactive percentage

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object