io.spokestack.spokestack.android

Class PreASRMicrophoneInput

java.lang.Object

io.spokestack.spokestack.android.PreASRMicrophoneInput

All Implemented Interfaces:

SpeechInput, AutoCloseable

public final class PreASRMicrophoneInput
extends Object
implements SpeechInput

A variant of MicrophoneInput that releases its internal AudioRecord when ASR is activated to avoid microphone conflicts.

This class uses the configured sample rate and always reads single-chanel 16-bit PCM samples.

Constructor Summary

Constructors

Constructor and Description
PreASRMicrophoneInput(SpeechConfig config) initializes a new microphone instance and opens the audio recorder.

Method Summary

Modifier and Type	Method and Description
void	close() Releases the resources associated with the microphone.
void	read(SpeechContext context, ByteBuffer frame) Reads a frame from the microphone if the pipeline is inactive.

Methods inherited from class java.lang.Object

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

Constructor Detail

PreASRMicrophoneInput

public PreASRMicrophoneInput(SpeechConfig config)

initializes a new microphone instance and opens the audio recorder.

Parameters:

config - speech pipeline configuration

Method Detail

close

public void close()

Releases the resources associated with the microphone.

Specified by:

close in interface AutoCloseable

read

public void read(SpeechContext context,
                 ByteBuffer frame)
          throws AudioRecordError

Reads a frame from the microphone if the pipeline is inactive. When the pipeline is activated, the microphone resource is released, and the speech context is flagged as being externally managed. The ASR component is expected to deactivate the pipeline to signal that Spokestack can recapture the microphone.

Specified by:

read in interface SpeechInput

Parameters:

context - the current speech context

frame - the frame buffer to fill

Throws:

AudioRecordError - if audio cannot be read