I have a wav file in my code. I used the FFT on my audio signal to do a STFT. Basically what I am doing is a "real-time spectral analysis". What I get from my signal is a two-sided spectrum.
The code here provided, helps you to understand, what I mean. I advise you to try it out. Just take any wav file and put it in your Matlab/Octave directory. The variable Yres
contains the the two-sided spectrum.
Any Idea, how to make it one-sided? I am sure it is something simple I am missing.
[y,fs]=wavread('UnchainMyHeart.wav');
t=linspace(0,length(y)/fs,length(y));
plot(t,y)
%Plotting my signal in the time domain
fftlen = 4096;
segl =floor(0.05*fs);
windowshift=segl/2;
window=hann(segl);
window=window.';
si=1;
ei=segl;
AOS= length(y)/windowshift - 1;
f1=figure;
f=0:1:fftlen-1;
f=f/(fftlen-1)*fs;
Ya=zeros(1,fftlen);
plot(f,Ya),axis([0 fs -90 20])
grid on
n=0;
for m = 1:1:AOS
y_a = y(si:ei);
y_a= y_a.*window;
Ya=fft(y_a, fftlen);
n=n+1;
if n==1
Yres=abs(Ya);
else
Yres=Yres+abs(Ya);
end
if n=10
Yres=Yres/10;
n=0;
drawnow;
figure(f1);
plot(f, 20*log10(abs(Yres)));
ylim([-90 20]);
title("Spectrum of a audio signal");
xlabel("f(Hz)");
ylabel("dB");
grid on;
end
si=si+windowshift;
ei=ei+windowshift;
end
you can just plot half of the spectrum you already have:
plot(f(1:end/2), 20*log10(abs(Yres(1:end/2))));
Or alternativly you can the full one but centered:
plot(fftshift(f), 20*log10(abs(fftshift(Yres))));