# tracking the pitch of audio signal. pls help

3 Ansichten (letzte 30 Tage)
danny Sharah am 9 Apr. 2015
Bearbeitet: danny Sharah am 9 Apr. 2015
i have audio signal in .wav format. its short time fourier transform analysis is done using below code and spectrogram plotted. now i want to plot its pitch contour. that is, want to track its pitch and study its variation over the time..can anybody help me with matlab code?
function spStftDemo [x, fs] = wavread('1.wav'); %x = wgn(1, 1000, 2); fs = 16000; frame_overlap = 10; % ms frame_length = 20; spStft(x(:), 'hamming', frame_overlap, frame_length, fs, 'plot');
function spStftDemo2
% use spectrogram (digital signal processing toolbox)
frame_overlap = 10; % ms
frame_length = 20;
window = 'rectwin';
nfft = round(frame_length * fs / 1000); % convert ms to points
noverlap = round(frame_overlap * fs / 1000); % convert ms to points
window = eval(sprintf('%s(nfft)', window)); % e.g., hamming(nfft)
[S, F, T, P] = spectrogram(x, window, noverlap, nfft, fs);
spectrogram(x, window, noverlap, nfft, fs, 'yaxis'); % plot
% NAME
% spStft: Short-Time Fourier Transform.
% SYNOPSIS
% [S, F, T] = spStft(x, [window], [frame_overlap], [frame_length],
% [fs], [show])
% DESCRIPTION
% Short-Time Fourier Transform
% INPUTS
% x (vector) of size Nx1.
% window (string) the window function such as rectwin, hamming.
% if not specified, equivalent to hamming
% frame_overlap
% (scalar) the length of each frame overlaps in micro second.
% The default is frame_length / 2.
% frame_length
% (scalar) the length of each frame in micro second.
% The default is 20ms.
% fs (scalar) the sampling rate in Hz. The default assumes 16kHz
% [show] (boolean) plot or not. The default is 0.
% OUTPUTS
% S (matrix) of size MxK where M is the size of each frame(fft))
% and K is the number of frames. S is spectrogram (complex number)
% F (vector) of size Mx1 that is a vector of frequencies (y-axis)
% in Hz.
% T (vector) of size 1xK whose value corresponds to the center of
% each frame (x-axis) in sec.
% AUTHOR
% Naotoshi Seo, April 2006
% spectrogram.m (digital signal processing toolbox)
% - arguments order was arranged similar to this (but in micro second).
function [S, F, T] = spStft(x, window, frame_overlap, frame_length, fs, show)
%%Initialization
N = length(x);
if ~exist('window', 'var') || isempty(window)
window = 'hamming';
end
if ~exist('fs', 'var') || isempty(fs)
fs = 16000;
end
if ~exist('frame_length', 'var') || isempty(frame_length)
frame_length = 20;
end
if ~exist('frame_overlap', 'var') || isempty(frame_overlap)
frame_overlap = frame_length / 2;
end
if ~exist('show', 'var') || isempty(show)
show = 0;
end
nsample = round(frame_length * fs / 1000); % convert ms to points
noverlap = round(frame_overlap * fs / 1000); % convert ms to points
window = eval(sprintf('%s(nsample)', window)); % e.g., hamming(nsample)
%%spectrogram
[S, F, T] = gen_spectrogram(x, window, noverlap, nsample, fs); % below
%%plot
if show
plot_spectrogram(S, F, T); % below
end
function [S, F, T] = gen_spectrogram(x, window, noverlap, nsample, fs)
N = length(x);
S = [];
pos = 1;
while (pos+nsample <= N)
frame = x(pos:pos+nsample-1);
pos = pos + (nsample - noverlap);
Y = fft(frame .* window, nsample);
S = [S Y(1:round(nsample/2), 1)]; % half is enough, another half is just mirror
end
[M, K] = size(S);
F = (0:round(nsample/2)-1)' / nsample * fs; % [0, fs/2) Hz
% F = psdfreqvec(nsample, fs, 'half');
T = (round(nsample/2):(nsample-noverlap):N-1-round(nsample/2))/fs;
% equals to spectrogram(..., 'yaxis'), no argout
function plot_spectrogram(S, F, T)
SdB = 20 * log10(abs(S)); % dB
ax = imagesc(T, F, SdB);
set(get(ax, 'Parent'), 'YDir', 'normal');
xlabel('Time')
ylabel('Frequency (Hz)')
% colorbar
##### 0 Kommentare-2 ältere Kommentare anzeigen-2 ältere Kommentare ausblenden

Melden Sie sich an, um zu kommentieren.

### Kategorien

Mehr zu Time-Frequency Analysis finden Sie in Help Center und File Exchange

### Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by