view dsp/mfcc/MFCC.h @ 321:f1e6be2de9a5

A threshold (delta) is added in the peak picking parameters structure (PPickParams). It is used as an offset when computing the smoothed detection function. A constructor for the structure PPickParams is also added to set the parameters to 0 when a structure instance is created. Hence programmes using the peak picking parameter structure and which do not set the delta parameter (e.g. QM Vamp note onset detector) won't be affected by the modifications. Functions modified: - dsp/onsets/PeakPicking.cpp - dsp/onsets/PeakPicking.h - dsp/signalconditioning/DFProcess.cpp - dsp/signalconditioning/DFProcess.h
author mathieub <mathieu.barthet@eecs.qmul.ac.uk>
date Mon, 20 Jun 2011 19:01:48 +0100
parents d5014ab8b0e5
children 701233f8ed41
line wrap: on
line source
/* -*- c-basic-offset: 4 indent-tabs-mode: nil -*-  vi:set ts=8 sts=4 sw=4: */

/*
    QM DSP Library

    Centre for Digital Music, Queen Mary, University of London.
    This file copyright 2005 Nicolas Chetry, copyright 2008 QMUL.

    This program is free software; you can redistribute it and/or
    modify it under the terms of the GNU General Public License as
    published by the Free Software Foundation; either version 2 of the
    License, or (at your option) any later version.  See the file
    COPYING included with this distribution for more information.
*/

#ifndef MFCC_H
#define MFCC_H

#include "base/Window.h"

class FFTReal;

struct MFCCConfig {
    int FS;
    int fftsize;
    int nceps;
    double logpower;
    bool want_c0;
    WindowType window;
    MFCCConfig(int _FS) :
        FS(_FS), fftsize(2048), nceps(19),
        logpower(1.0), want_c0(true), window(HammingWindow) { }
};

class MFCC
{
public:
    MFCC(MFCCConfig config);
    virtual ~MFCC();

    /**
     * Process time-domain input data.  inframe must contain
     * getfftlength() samples.  outceps must contain space for nceps
     * values, plus one if want_c0 is specified.
     */
    int process(const double *inframe, double *outceps);

    /**
     * Process time-domain input data.  real and imag must contain
     * getfftlength()/2+1 elements (i.e. the conjugate half of the FFT
     * is not expected).  outceps must contain space for nceps values,
     * plus one if want_c0 is specified.
     */
    int process(const double *real, const double *imag, double *outceps);

    int getfftlength() const { return fftSize; }

private:
    /* Filter bank parameters */
    double  lowestFrequency; 
    int     linearFilters; 
    double  linearSpacing;
    int     logFilters;
    double  logSpacing;
    
    /* FFT length */
    int     fftSize;
    
    int     totalFilters;
    double  logPower;
    
    /* Misc. */
    int     samplingRate;
    int     nceps;
    
    /* MFCC vector */
    double  *ceps;
    
    double  **mfccDCTMatrix;
    double  **mfccFilterWeights;
    
    /* The analysis window */
    Window<double> *window;
    
    /* For the FFT */
    double *realOut;
    double *imagOut;
    double *fftMag;
    double *earMag;
    FFTReal *fft;

    /* Set if user want C0 */
    int WANT_C0;
};


#endif