-
Notifications
You must be signed in to change notification settings - Fork 0
/
FeatureSpectralSkewness.m
executable file
·35 lines (30 loc) · 1.28 KB
/
FeatureSpectralSkewness.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
% ======================================================================
%> @brief computes the spectral skewness from the magnitude spectrum
%> called by ::ComputeFeature
%>
%> @param X: spectrogram (dimension FFTLength X Observations)
%> @param f_s: sample rate of audio data (unused)
%>
%> @retval v spectral skewness
% ======================================================================
function [vssk] = FeatureSpectralSkewness (X, f_s)
UseBookDefinition = true;
if (UseBookDefinition)
% compute mean and standard deviation
mu_x = mean(abs(X), 1);
std_x = std(abs(X), 1);
% compute skewness
X = X - repmat(mu_x, size(X,1), 1);
vssk = sum ((X.^3)./(repmat(std_x, size(X,1), 1).^3*size(X,1)));
else
% interpret the spectrum as pdf, not as signal
f = linspace(0, f_s/2, size(X,1));
% compute mean and standard deviation
mu_X = (f * X) ./ (sum(X,1));
tmp = repmat(f, size(X,2),1) - repmat(mu_X, size(X,1),1)';
var_X = diag (tmp.^2 * X) ./ (sum(X,1)'*size(X,1));
vssk = diag (tmp.^3 * X) ./ (var_X.^(3/2) .* sum(X,1)'*size(X,1));
end
% avoid NaN for silence frames
vssk (sum(X,1) == 0) = 0;
end