marytts.vocalizations
Class MLSASynthesisTechnology

java.lang.Object
  extended by marytts.vocalizations.VocalizationSynthesisTechnology
      extended by marytts.vocalizations.MLSASynthesisTechnology

public class MLSASynthesisTechnology
extends VocalizationSynthesisTechnology

MLSA Synthesis technology to synthesize vocalizations

Author:
Sathish Pammi

Field Summary
protected  HMMData htsData
           
protected  boolean imposePolynomialContour
           
protected  HTSVocoder par2speech
           
protected  VocalizationIntonationReader vIntonationReader
           
protected  MLSAFeatureFileReader vMLSAFeaturesReader
           
 
Constructor Summary
MLSASynthesisTechnology(java.lang.String mlsaFeatureFile, java.lang.String intonationFeatureFile, java.lang.String mixedExcitationFile, boolean imposePolynomialContour)
           
 
Method Summary
 javax.sound.sampled.AudioInputStream reSynthesize(int backchannelNumber, javax.sound.sampled.AudioFileFormat aft)
          Re-synthesize given vocalization using MLSA (it is same as synthesize())
 javax.sound.sampled.AudioInputStream synthesize(int backchannelNumber, javax.sound.sampled.AudioFileFormat aft)
          Synthesize given vocalization using MLSA vocoder
 javax.sound.sampled.AudioInputStream synthesizeUsingImposedF0(int sourceIndex, int targetIndex, javax.sound.sampled.AudioFileFormat aft)
          Impose target intonation contour on given vocalization using MLSA technology
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Field Detail

vMLSAFeaturesReader

protected MLSAFeatureFileReader vMLSAFeaturesReader

vIntonationReader

protected VocalizationIntonationReader vIntonationReader

htsData

protected HMMData htsData

par2speech

protected HTSVocoder par2speech

imposePolynomialContour

protected boolean imposePolynomialContour
Constructor Detail

MLSASynthesisTechnology

public MLSASynthesisTechnology(java.lang.String mlsaFeatureFile,
                               java.lang.String intonationFeatureFile,
                               java.lang.String mixedExcitationFile,
                               boolean imposePolynomialContour)
                        throws MaryConfigurationException
Throws:
MaryConfigurationException
Method Detail

synthesize

public javax.sound.sampled.AudioInputStream synthesize(int backchannelNumber,
                                                       javax.sound.sampled.AudioFileFormat aft)
                                                throws SynthesisException
Synthesize given vocalization using MLSA vocoder

Specified by:
synthesize in class VocalizationSynthesisTechnology
Parameters:
unitIndex - unit index
aft - audio file format
Returns:
AudioInputStream of synthesized vocalization
Throws:
SynthesisException - if failed to synthesize vocalization

reSynthesize

public javax.sound.sampled.AudioInputStream reSynthesize(int backchannelNumber,
                                                         javax.sound.sampled.AudioFileFormat aft)
                                                  throws SynthesisException
Re-synthesize given vocalization using MLSA (it is same as synthesize())

Specified by:
reSynthesize in class VocalizationSynthesisTechnology
Parameters:
unitIndex - unit index
aft - audio file format
Returns:
AudioInputStream of synthesized vocalization
Throws:
SynthesisException - if failed to synthesize vocalization

synthesizeUsingImposedF0

public javax.sound.sampled.AudioInputStream synthesizeUsingImposedF0(int sourceIndex,
                                                                     int targetIndex,
                                                                     javax.sound.sampled.AudioFileFormat aft)
                                                              throws SynthesisException
Impose target intonation contour on given vocalization using MLSA technology

Specified by:
synthesizeUsingImposedF0 in class VocalizationSynthesisTechnology
Parameters:
sourceIndex - unit index of vocalization
targetIndex - unit index of target intonation
aft - audio file format
Returns:
AudioInputStream of synthesized vocalization
Throws:
SynthesisException - if failed to synthesize vocalization