org.tensorflow.op.audio

Class Mfcc

java.lang.Object

org.tensorflow.op.PrimitiveOp

org.tensorflow.op.audio.Mfcc

All Implemented Interfaces:

Op, Operand<Float>

@Operator(group="audio")
public final class Mfcc
extends PrimitiveOp
implements Operand<Float>

Transforms a spectrogram into a form that's useful for speech recognition.

Mel Frequency Cepstral Coefficients are a way of representing audio data that's been effective as an input feature for machine learning. They are created by taking the spectrum of a spectrogram (a 'cepstrum'), and discarding some of the higher frequencies that are less significant to the human ear. They have a long history in the speech recognition world, and https://en.wikipedia.org/wiki/Mel-frequency_cepstrum is a good resource to learn more.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	Mfcc.Options Optional attributes for Mfcc

Field Summary

Fields inherited from class org.tensorflow.op.PrimitiveOp

operation

Method Summary

Modifier and Type	Method and Description
Output<Float>	asOutput() Returns the symbolic handle of a tensor.
static Mfcc	create(Scope scope, Operand<Float> spectrogram, Operand<Integer> sampleRate, Mfcc.Options... options) Factory method to create a class wrapping a new Mfcc operation.
static Mfcc.Options	dctCoefficientCount(Long dctCoefficientCount)
static Mfcc.Options	filterbankChannelCount(Long filterbankChannelCount)
static Mfcc.Options	lowerFrequencyLimit(Float lowerFrequencyLimit)
Output<Float>	output()
static Mfcc.Options	upperFrequencyLimit(Float upperFrequencyLimit)

Methods inherited from class org.tensorflow.op.PrimitiveOp

equals, hashCode, op, toString

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Method Detail

create

public static Mfcc create(Scope scope,
                          Operand<Float> spectrogram,
                          Operand<Integer> sampleRate,
                          Mfcc.Options... options)

Factory method to create a class wrapping a new Mfcc operation.

Parameters:

scope - current scope

spectrogram - Typically produced by the Spectrogram op, with magnitude_squared set to true.

sampleRate - How many samples per second the source audio used.

options - carries optional attributes values

Returns:

a new instance of Mfcc

upperFrequencyLimit

public static Mfcc.Options upperFrequencyLimit(Float upperFrequencyLimit)

Parameters:

upperFrequencyLimit - The highest frequency to use when calculating the ceptstrum.

lowerFrequencyLimit

public static Mfcc.Options lowerFrequencyLimit(Float lowerFrequencyLimit)

Parameters:

lowerFrequencyLimit - The lowest frequency to use when calculating the ceptstrum.

filterbankChannelCount

public static Mfcc.Options filterbankChannelCount(Long filterbankChannelCount)

Parameters:

filterbankChannelCount - Resolution of the Mel bank used internally.

dctCoefficientCount

public static Mfcc.Options dctCoefficientCount(Long dctCoefficientCount)

Parameters:

dctCoefficientCount - How many output channels to produce per time slice.

output

public Output<Float> output()

asOutput

public Output<Float> asOutput()

Description copied from interface: Operand

Returns the symbolic handle of a tensor.

Inputs to TensorFlow operations are outputs of another TensorFlow operation. This method is used to obtain a symbolic handle that represents the computation of the input.

Specified by:

asOutput in interface Operand<Float>

See Also:

OperationBuilder.addInput(Output)