javaCheetah - Java API

API Reference for the Java Cheetah SDK (cheetah-java)

package: ai.picovoice.cheetah


Cheetah

public class Cheetah { }

Class for the Cheetah Speech-to-Text engine.

Cheetah can be initialized using the Builder() Class. Resources should be cleaned when you are done using the delete() function.


Cheetah.Cheetah()

public Cheetah(
String accessKey,
String libraryPath,
String modelPath,
float endpointDurationSec) throws CheetahException

Cheetah constructor. See also the Cheetah.Builder Class.

Parameters

  • accessKey String : AccessKey obtained from Picovoice Console.
  • libraryPath String : Absolute path to the native Cheetah library.
  • modelPath String : Absolute path to the file containing model parameters.
  • endpointDurationSec float : Duration of endpoint in seconds. A speech endpoint is detected when there is a chunk of audio (with a duration specified herein) after an utterance without any speech in it. Set duration to 0 to disable this. Default is 1 second in the Builder.

Returns

  • Cheetah: An instance of Cheetah engine.

Throws


Cheetah.delete()

public void delete()

Releases resources acquired by Cheetah.


Cheetah.getFrameLength()

public int getFrameLength()

Getter for number of audio samples per frame.

Returns

  • int: Number of audio samples per frame.

Cheetah.getSampleRate()

public int getSampleRate()

Getter for required audio sample rate for PCM data.

Returns

  • int: Required audio sample rate for PCM data.

Cheetah.getVersion()

public String getVersion()

Getter for version.

Returns


Cheetah.process()

public CheetahTranscript process(short[] pcm) throws CheetahException

Processes a frame of audio and returns newly-transcribed text and a flag indicating if an endpoint has been detected. Upon detection of an endpoint, the client may invoke .flush() to retrieve any remaining transcription.

The number of samples per frame can be attained by calling .getFrameLength(). The incoming audio needs to have a sample rate equal to .getSampleRate() and be 16-bit linearly-encoded. Furthermore, Cheetah operates on single-channel audio.

Parameters

  • pcm short[] : A frame of audio samples.

Returns

Throws


Cheetah.flush()

public CheetahTranscript flush() throws CheetahException

Processes any remaining audio data and returns its transcription.

Returns

Throws


Cheetah.Builder

public static class Builder {
public Builder(String accessKey) { }
}

Builder for creating an instance of Cheetah with a mixture of default arguments.

Parameters


Cheetah.Builder.build()

public Cheetah build() throws CheetahException

Creates an instance of Cheetah Speech-to-Text engine.

Returns

  • Cheetah: An instance of Cheetah Speech-to-Text engine.

Throws

  • CheetahException: If an error occurs while creating an instance of Cheetah Speech-to-Text engine.

Cheetah.Builder.setAccessKey()

public Cheetah.Builder setAccessKey(String accessKey)

Sets the AccessKey of the builder.

Parameters

Returns


Cheetah.Builder.setModelPath()

public Cheetah.Builder setModelPath(String modelPath)

Sets the model path of the builder.

Parameters

  • modelPath String : Absolute path to the file containing model parameters (.pv).

Returns


Cheetah.Builder.setLibraryPath()

public Builder setLibraryPath(String libraryPath)

Sets the library path of the builder.

Parameters

  • libraryPath String : Absolute path to the native Cheetah library.

Returns


Cheetah.Builder.setEndpointDuration()

public Cheetah.Builder setEndpointDuration(float endpointDuration)

Sets the endpoint duration of the builder. A speech endpoint is detected when there is a chunk of audio (with a duration specified herein) after an utterance without any speech in it. Set duration to 0 to disable this. Default is 1 second.

Parameters

  • endpointDuration float : Duration of endpoint in seconds.

Returns


CheetahTranscript

public class CheetahTranscript { }

Class that contains Cheetah transcript data.


CheetahTranscript.getTranscript()

public string getTranscript() { }

Getter for transcript data.

Returns

  • string: Inferred transcription.

CheetahTranscript.getIsEndpoint()

public boolean getIsEndpoint() { }

Getter for isEndpoint flag.

Returns

  • boolean: If true, Cheetah detected a speech endpoint.

CheetahException

public class CheetahException extends Exception { }

Exception thrown if an error occurs within Cheetah Speech-to-Text engine.

Exceptions:

public class CheetahActivationException extends CheetahException { }
public class CheetahActivationLimitException extends CheetahException { }
public class CheetahActivationRefusedException extends CheetahException { }
public class CheetahActivationThrottledException extends CheetahException { }
public class CheetahIOException extends CheetahException { }
public class CheetahInvalidArgumentException extends CheetahException { }
public class CheetahInvalidStateException extends CheetahException { }
public class CheetahKeyException extends CheetahException { }
public class CheetahMemoryException extends CheetahException { }
public class CheetahRuntimeException extends CheetahException { }
public class CheetahStopIterationException extends CheetahException { }

Issue with this doc? Please let us know.