Hi , I make spectral analysis for female voice ( wav file for 2 word only ) , I get frequency contents below 1000 Hz , so is this correct , also I suspect computation of power of abs(fft) , please advise , thanks

file1='Left_2_f.wav';
[y,fs] = audioread(file1);
L=length(y);
t=(1000*(0:1/fs:(L-1)/fs))';
n = 2^nextpow2(L);
Y=fft(y,n);
P = abs(Y/n);
P = P(1:n/2+1);
P(2:end-1) = 2*P(2:end-1);
power = abs(Y).^2/n;    % power of the DFT
power = power(1:n/2+1);
power(2:end-1) = 2*power(2:end-1);
f = fs*(0:(n/2))/n;
f1 = fs*(0:n-1)/n;
figure
subplot(4,1,1)
plot(t,y)
xlabel('Time in mseconds');
ylabel('Audio Level');
title('Plot of female voice ');
grid on
grid minor
subplot(4,1,2)
plot(f,P) 
title('Single-Sided Magnitude Spectrum of female voice ')
xlabel('f (Hz)')
ylabel('Magnitude')
grid on
grid minor
subplot(4,1,3)
[pk,j] = findpeaks(P,'SortStr','descend','NPeaks',8);
Peaks(:,j) = f(j);
xx=(f(:,j));
xx=(sort(xx));
[~,loc] = max(P);
plot(f(1:10*loc),P(1:10*loc),f(j),pk,'o') 
title(['The maixmum of Magnitude Spectrum occurs at ( ' ,num2str(xx) , ' ) Hz' ])
xlabel('f (Hz)')
ylabel('Magnitude')
grid on
grid minor
subplot(4,1,4)
plot(f,power)
xlabel('Frequency')
ylabel('Power')

Answer 1

[Image Analyst]

It's correct - females can have audio power below 1 kHz.

To get power, try pwelch() or bandpower().

[Yasir Muhammad]

would uu tell me how to read all audio files and then ploting fft and maximum spectrum

[Image Analyst]

See the FAQ:

https://matlab.fandom.com/wiki/FAQ#How_can_I_process_a_sequence_of_files.3F