edu.cmu.sphinx.frontend.transform

Class DiscreteCosineTransform

java.lang.Object

edu.cmu.sphinx.util.props.ConfigurableAdapter

edu.cmu.sphinx.frontend.BaseDataProcessor

edu.cmu.sphinx.frontend.transform.DiscreteCosineTransform

All Implemented Interfaces:

DataProcessor, Configurable

Direct Known Subclasses:

DiscreteCosineTransform2, KaldiDiscreteCosineTransform

public class DiscreteCosineTransform
extends BaseDataProcessor

Applies a logarithm and then a Discrete Cosine Transform (DCT) to the input data. The input data is normally the mel spectrum. It has been proven that, for a sequence of real numbers, the discrete cosine transform is equivalent to the discrete Fourier transform. Therefore, this class corresponds to the last stage of converting a signal to cepstra, defined as the inverse Fourier transform of the logarithm of the Fourier transform of a signal. The property PROP_CEPSTRUM_LENGTHrefers to the dimensionality of the coefficients that are actually returned, defaulting to 13. When the input is mel-spectrum, the vector returned is the MFCC (Mel-Frequency Cepstral Coefficient) vector, where the 0-th element is the energy value.

Field Summary

Fields

Modifier and Type	Field and Description
protected int	cepstrumSize
protected double[][]	melcosine
protected int	numberMelFilters
static java.lang.String	PROP_CEPSTRUM_LENGTH The property for the size of the cepstrum
static java.lang.String	PROP_NUMBER_FILTERS The property for the number of filters in the filterbank.

Fields inherited from class edu.cmu.sphinx.util.props.ConfigurableAdapter

logger

Constructor Summary

Constructors

Constructor and Description
DiscreteCosineTransform()
DiscreteCosineTransform(int numberMelFilters, int cepstrumSize)

Method Summary

Modifier and Type	Method and Description
protected double[]	applyMelCosine(double[] melspectrum) Apply the MelCosine filter to the given melspectrum.
protected void	computeMelCosine() Compute the MelCosine filter bank.
Data	getData() Returns the next DoubleData object, which is the mel cepstrum of the input frame.
void	initialize() Initializes this DataProcessor.
void	newProperties(PropertySheet ps) This method is called when this configurable component needs to be reconfigured.

Methods inherited from class edu.cmu.sphinx.frontend.BaseDataProcessor

getPredecessor, setPredecessor

Methods inherited from class edu.cmu.sphinx.util.props.ConfigurableAdapter

getName, initLogger, toString

Methods inherited from class java.lang.Object

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

Field Detail

PROP_NUMBER_FILTERS

@S4Integer(defaultValue=40)
public static final java.lang.String PROP_NUMBER_FILTERS

The property for the number of filters in the filterbank.

See Also:

Constant Field Values

PROP_CEPSTRUM_LENGTH

@S4Integer(defaultValue=13)
public static final java.lang.String PROP_CEPSTRUM_LENGTH

The property for the size of the cepstrum

See Also:

Constant Field Values

cepstrumSize

protected int cepstrumSize

numberMelFilters

protected int numberMelFilters

melcosine

protected double[][] melcosine

Constructor Detail

DiscreteCosineTransform

public DiscreteCosineTransform(int numberMelFilters,
                               int cepstrumSize)

DiscreteCosineTransform

public DiscreteCosineTransform()

Method Detail

newProperties

public void newProperties(PropertySheet ps)
                   throws PropertyException

Description copied from interface: Configurable

This method is called when this configurable component needs to be reconfigured.

Specified by:

newProperties in interface Configurable

Overrides:

newProperties in class ConfigurableAdapter

Parameters:

ps - a property sheet holding the new data

Throws:

PropertyException - if there is a problem with the properties.

initialize

public void initialize()

Description copied from class: BaseDataProcessor

Initializes this DataProcessor. This is typically called after the DataProcessor has been configured.

Specified by:

initialize in interface DataProcessor

Overrides:

initialize in class BaseDataProcessor

getData

public Data getData()
             throws DataProcessingException

Returns the next DoubleData object, which is the mel cepstrum of the input frame. Signals are returned unmodified.

Specified by:

getData in interface DataProcessor

Specified by:

getData in class BaseDataProcessor

Returns:

the next available DoubleData melcepstrum, or Signal object, or null if no Data is available

Throws:

DataProcessingException - if a data processing error occurred

computeMelCosine

protected void computeMelCosine()

Compute the MelCosine filter bank.

applyMelCosine

protected double[] applyMelCosine(double[] melspectrum)

Apply the MelCosine filter to the given melspectrum.

Parameters:

melspectrum - the MelSpectrum data

Returns:

MelCepstrum data produced by apply the MelCosine filter to the MelSpectrum data