marytts.signalproc.sinusoidal

Class SinusoidalAnalyzer

java.lang.Object

marytts.signalproc.sinusoidal.BaseSinusoidalAnalyzer

marytts.signalproc.sinusoidal.SinusoidalAnalyzer

Direct Known Subclasses:

PitchSynchronousSinusoidalAnalyzer

public class SinusoidalAnalyzer
extends BaseSinusoidalAnalyzer

Sinusoidal Modeling Analysis Module Given a speech/audio signal, a set of amplitudes, frequencies and phases are estimated on a frame-by-frame basis Then, sinusoids that are close in frequency are grouped together to form sinusoidal tracks Optional amplitude and phase continuity constraints can be employed during track generation The implementation consists of ideas and algorithms from various papers as described in function headers References: Quatieri, T. F. Discrete-Time Speech Signal Processing: Principles and Practice. Prentice-Hall Inc. 2001. (Chapter 9 – Sinusoidal Analysis/Synthesis) R.J. McAulay and T.F. Quatieri, "Speech Analysis/Synthesis Based on a Sinusoidal Representation," IEEE Transactions on Acoustics, Speech and Signal Processing, vol. ASSP-34, no. 4, August 1986.

Author:

Oytun Türk

Field Summary

Fields

Modifier and Type	Field and Description
SinusoidalAnalysisParams	params

Fields inherited from class marytts.signalproc.sinusoidal.BaseSinusoidalAnalyzer

FIXEDRATE_FULLBAND_ANALYZER, FIXEDRATE_MULTIRESOLUTION_ANALYZER, PITCHSYNCHRONOUS_FULLBAND_ANALYZER, PITCHSYNCHRONOUS_MULTIRESOLUTION_ANALYZER

Constructor Summary

Constructors

Constructor and Description
SinusoidalAnalyzer(SinusoidalAnalysisParams paramsIn)

Method Summary

Methods

Modifier and Type	Method and Description
static NonharmonicSinusoidalSpeechFrame	analyze_frame(double[] frm, boolean isOutputToTextFile, boolean isVoiced, SinusoidalAnalysisParams params)
static NonharmonicSinusoidalSpeechFrame	analyze_frame(double[] windowedFrm, boolean isOutputToTextFile, int spectralEnvelopeType, boolean isVoiced, float f0, float maxFreqOfVoicing, boolean bEstimateHNMVoicing, SinusoidalAnalysisParams params, float[] initialPeakLocationsInHz)
static NonharmonicSinusoidalSpeechFrame	analyze_frame(double[] frm, boolean isOutputToTextFile, int spectralEnvelopeType, boolean isVoiced, float f0, SinusoidalAnalysisParams params)
static NonharmonicSinusoidalSpeechFrame	analyze_frame(double[] frm, boolean isOutputToTextFile, int spectralEnvelopeType, boolean isVoiced, SinusoidalAnalysisParams params)
SinusoidalTracks	analyzeFixedRate(double[] x)
SinusoidalTracks	analyzeFixedRate(double[] x, float winSizeInSeconds)
SinusoidalTracks	analyzeFixedRate(double[] x, float winSizeInSeconds, float skipSizeInSeconds)
SinusoidalTracks	analyzeFixedRate(double[] x, float winSizeInSeconds, float skipSizeInSeconds, float deltaInHz)
SinusoidalTracks	analyzeFixedRate(double[] x, float winSizeInSeconds, float skipSizeInSeconds, float deltaInHz, int spectralEnvelopeType)
SinusoidalTracks	analyzeFixedRate(double[] x, float winSizeInSeconds, float skipSizeInSeconds, float deltaInHz, int spectralEnvelopeType, double[] f0s, float ws_f0, float ss_f0)
NonharmonicSinusoidalSpeechSignal	extractSinusoidsFixedRate(double[] x, float winSizeInSeconds, float skipSizeInSeconds, float deltaInHz)
NonharmonicSinusoidalSpeechSignal	extractSinusoidsFixedRate(double[] x, float winSizeInSeconds, float skipSizeInSeconds, float deltaInHz, int spectralEnvelopeType)
NonharmonicSinusoidalSpeechSignal	extractSinusoidsFixedRate(double[] x, float winSizeInSeconds, float skipSizeInSeconds, float deltaInHz, int spectralEnvelopeType, double[] f0s, float ws_f0, float ss_f0)
static void	getGrossStatistics(SinusoidalTracks sinTracks)
static void	main(String[] args)
static float[]	refineBySpectralReassignment(ComplexArray windowedFrameFFT, ComplexArray windowedFrameDerivativeFFT, float[] freqInds, int fftSize, int fs)
static FreqIndicesWithAmplitudes	refinePeakEstimatesBias(double[] powSpecdB, float[] freqInds, int windowType)
static FreqIndicesWithAmplitudes	refinePeakEstimatesParabola(double[] powSpecdB, int[] freqInds)
static int[]	setNeighFreq(int fftSize, boolean bAdjustNeighFreqDependent, float fs)

Methods inherited from class java.lang.Object

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

Field Detail

params

public SinusoidalAnalysisParams params

Constructor Detail

SinusoidalAnalyzer

public SinusoidalAnalyzer(SinusoidalAnalysisParams paramsIn)

Method Detail

setNeighFreq

public static int[] setNeighFreq(int fftSize,
                 boolean bAdjustNeighFreqDependent,
                 float fs)

analyzeFixedRate

public SinusoidalTracks analyzeFixedRate(double[] x)

analyzeFixedRate

public SinusoidalTracks analyzeFixedRate(double[] x,
                                float winSizeInSeconds)

analyzeFixedRate

public SinusoidalTracks analyzeFixedRate(double[] x,
                                float winSizeInSeconds,
                                float skipSizeInSeconds)

analyzeFixedRate

public SinusoidalTracks analyzeFixedRate(double[] x,
                                float winSizeInSeconds,
                                float skipSizeInSeconds,
                                float deltaInHz)

analyzeFixedRate

public SinusoidalTracks analyzeFixedRate(double[] x,
                                float winSizeInSeconds,
                                float skipSizeInSeconds,
                                float deltaInHz,
                                int spectralEnvelopeType)

analyzeFixedRate

public SinusoidalTracks analyzeFixedRate(double[] x,
                                float winSizeInSeconds,
                                float skipSizeInSeconds,
                                float deltaInHz,
                                int spectralEnvelopeType,
                                double[] f0s,
                                float ws_f0,
                                float ss_f0)

extractSinusoidsFixedRate

public NonharmonicSinusoidalSpeechSignal extractSinusoidsFixedRate(double[] x,
                                                          float winSizeInSeconds,
                                                          float skipSizeInSeconds,
                                                          float deltaInHz)

extractSinusoidsFixedRate

public NonharmonicSinusoidalSpeechSignal extractSinusoidsFixedRate(double[] x,
                                                          float winSizeInSeconds,
                                                          float skipSizeInSeconds,
                                                          float deltaInHz,
                                                          int spectralEnvelopeType)

extractSinusoidsFixedRate

public NonharmonicSinusoidalSpeechSignal extractSinusoidsFixedRate(double[] x,
                                                          float winSizeInSeconds,
                                                          float skipSizeInSeconds,
                                                          float deltaInHz,
                                                          int spectralEnvelopeType,
                                                          double[] f0s,
                                                          float ws_f0,
                                                          float ss_f0)

getGrossStatistics

public static void getGrossStatistics(SinusoidalTracks sinTracks)

analyze_frame

public static NonharmonicSinusoidalSpeechFrame analyze_frame(double[] frm,
                                             boolean isOutputToTextFile,
                                             boolean isVoiced,
                                             SinusoidalAnalysisParams params)

analyze_frame

public static NonharmonicSinusoidalSpeechFrame analyze_frame(double[] frm,
                                             boolean isOutputToTextFile,
                                             int spectralEnvelopeType,
                                             boolean isVoiced,
                                             SinusoidalAnalysisParams params)

analyze_frame

public static NonharmonicSinusoidalSpeechFrame analyze_frame(double[] frm,
                                             boolean isOutputToTextFile,
                                             int spectralEnvelopeType,
                                             boolean isVoiced,
                                             float f0,
                                             SinusoidalAnalysisParams params)

analyze_frame

public static NonharmonicSinusoidalSpeechFrame analyze_frame(double[] windowedFrm,
                                             boolean isOutputToTextFile,
                                             int spectralEnvelopeType,
                                             boolean isVoiced,
                                             float f0,
                                             float maxFreqOfVoicing,
                                             boolean bEstimateHNMVoicing,
                                             SinusoidalAnalysisParams params,
                                             float[] initialPeakLocationsInHz)

refinePeakEstimatesParabola

public static FreqIndicesWithAmplitudes refinePeakEstimatesParabola(double[] powSpecdB,
                                                    int[] freqInds)

refinePeakEstimatesBias

public static FreqIndicesWithAmplitudes refinePeakEstimatesBias(double[] powSpecdB,
                                                float[] freqInds,
                                                int windowType)

refineBySpectralReassignment

public static float[] refineBySpectralReassignment(ComplexArray windowedFrameFFT,
                                   ComplexArray windowedFrameDerivativeFFT,
                                   float[] freqInds,
                                   int fftSize,
                                   int fs)

main

public static void main(String[] args)
                 throws UnsupportedAudioFileException,
                        IOException

Throws:

UnsupportedAudioFileException

IOException