Matlab mfcc函数参数详解
其实可以直接打开源代码看哈。
%MFCC Extract the mfcc, log-energy, delta, and delta-delta of audio signal
% coeffs = MFCC(audioIn,fs) returns the mel-frequency cepstral
% coefficients over time for the audio input. Columns of the input are
% treated as individual channels. coeffs is returned as an L-by-M-by-N
% array.
% L - Number of frames the audio signal is partitioned into.
% This is determined by the WINDOWLENGTH and OVERLAPLENGTH
% properties.
% M - Number coefficients returned per frame.
% This is determined by the NUMCOEFFS property.
% N - Number of channels.
%
% coeffs = MFCC(...,'WindowLength',WINDOWLENGTH) specifies the analysis
% window length used to calculate the coefficients. Specify the window
% length in samples as a positive scalar. If unspecified, WINDOWLENGTH
% defaults to round(0.030 * fs).
%
% coeffs = MFCC(...,'OverlapLength',OVERLAPLENGTH) specifies the number
% of samples overlap between adjacent windows. Specify the overlap length
% as a scalar smaller than the window length. If unspecified,
% OVERLAPLENGTH defaults to round(fs*0.02).
%
% coeffs = MFCC(...,'NumCoeffs',NUMCOEFFS) specifies the number
% of coefficients returned for each window. If not specified, the number
% of coefficients is 13.
%
% coeffs = MFCC(...,'FFTLength',FFTLENGTH) specifies the FFT length. By
% default, the FFT length is set to the WINDOWLENGTH.
%
% coeffs = MFCC(...,'DeltaWindowLength',DELTAWINDOWLENGTH) specifies the
% delta and delta-delta window length. The default is 2.
%
% coeffs = MFCC(...,'LogEnergy',LOGENERGY) specifies if and how the log
% energy is used. Specify log energy as a character vector:
% 'Append' - Adds the log-energy as the first element of the
% returned coefficients vector. This is the default
% setting.
% 'Replace' - Replaces the zeroth coefficient (first element of
% coeffs) with the log-energy.
% 'Ignore' - Ignores and does not return the log-energy.
%
% [coeffs,delta,deltaDelta,loc] = MFCC(...) returns the delta,
% delta-delta, and location in samples of each window of data. The
% location is the most recent sample used to calculate the coefficients.
% The delta and delta-delta vectors are derived from how the coeffs
% vector was specified. For example, if 'LogEnergy' is set to 'Replace',
% the first element of the coeffs vector is the log-energy, and the first
% element of the delta vector is the delta of the log-energy.
%
% EXAMPLE 1: Get the mel-frequency cepstral coefficients for entire
% speech file
% % Use the default settings to extract the MFCC feature from a speech
% % file.
%
% [audioIn,fs] = audioread('Counting-16-44p1-mono-15secs.wav');
% [coeffs,delta,deltaDelta,loc] = mfcc(audioIn,fs);