how to change this code to plot the frequency response of the desired interval?

Hi! I am using the R2021b Matlab and I need this code to plot the frequency response of the audio in the specified time interval. Could anyone help me please?

The code is displayed below:

recObj = audiorecorder;
Fs=8000;
filename = sprintf('myAudioData.wav');
disp('Start speaking.')
recordblocking(recObj, 10);
disp('End of Recording.');

doubleArray = getaudiodata(recObj);
audiowrite(filename,doubleArray,Fs);


[x,Fs] = audioread('myAudioData.wav');
ts = 1 / Fs;
n = round(0.02 / ts); 
nexttile       
t1 = 1.000; 
t2 = 1.020; 
% start and end index in audio data of segment of interest
idx1 = round(t1/ts);
idx2 = round(t2/ts);
% plot it!
plot(x(idx1:idx2));       % plot 1st 20 ms of audio file

nexttile
t3 = 3.000; 
t4 = 3.020; 
% start and end index in audio data of segment of interest
idx3 = round(t3/ts);
idx4 = round(t4/ts);
% plot it!
plot(x(idx3:idx4));

Answer 1

[Star Strider]

That depends on what you want as the result.

Two possibilities are fft and pspectrum with the 'spectrogram' option.

[Michael Sugiarto]

Probably ffts are better

[Michael Sugiarto]

could you give me an example on how to use it please?

[Star Strider]

Sure!

load handel.mat

filename = 'handel.wav';
audiowrite(filename,y,Fs);
clear y Fs

[x,Fs] = audioread('handel.wav');
t = linspace(0, size(x,1)-1, size(x,1))/Fs;
L = numel(t);

figure
plot(t, x)
grid

Fn = Fs/2;
NFFT = 2^nextpow2(L);
FTx = fft(x - mean(x), NFFT) / L;
Fv = linspace(0, 1, NFFT/2+1)*Fn;
Iv = 1:numel(Fv);

figure
plot(Fv, abs(FTx(Iv))*2)
grid
xlim([min(Fv) max(Fv)])
title('Fast Fourier Transform')

figure
pspectrum(x,Fs, 'spectrogram')
colormap(turbo)

.

[Michael Sugiarto]

But, how do I fft with the time intervals that I need?

[Star Strider]

Try this —

load handel.mat

filename = 'handel.wav';
audiowrite(filename,y,Fs);
clear y Fs

tiledlayout(2,3)
[x,Fs] = audioread('handel.wav');
t = linspace(0, size(x,1)-1, size(x,1))/Fs;
L = numel(t);
ts = 1 / Fs;

nexttile   
t1 = 1.000; 
t2 = 1.020; 
% start and end index in audio data of segment of interest
idx1 = round(t1/ts);
idx2 = round(t2/ts);
% plot it!
x1 = x(idx1:idx2);
plot(x(idx1:idx2));       % plot 1st 20 ms of audio file
title('x_1')

L = numel(x1);
Fn = Fs/2;
NFFT = 2^nextpow2(L);
FTx1 = fft(x1 - mean(x1), NFFT) / L;
Fv1 = linspace(0, 1, NFFT/2+1)*Fn;
Iv1 = 1:numel(Fv1);

nexttile
plot(Fv1, abs(FTx1(Iv1))*2)
grid
xlim([min(Fv1) max(Fv1)])
title('FFT')

nexttile
pspectrum(x1,Fs, 'spectrogram')
colormap(turbo)


nexttile
t3 = 3.000; 
t4 = 3.020; 
% start and end index in audio data of segment of interest
idx3 = round(t3/ts);
idx4 = round(t4/ts);
% plot it!
x2 = x(idx3:idx4);
plot(x(idx3:idx4));
title('x_2')

L = numel(x2);
Fn = Fs/2;
NFFT = 2^nextpow2(L);
FTx2 = fft(x2 - mean(x2), NFFT) / L;
Fv2 = linspace(0, 1, NFFT/2+1)*Fn;
Iv2 = 1:numel(Fv2);

nexttile
plot(Fv2, abs(FTx2(Iv2))*2)
grid
xlim([min(Fv2) max(Fv2)])
title('FFT')

nexttile
pspectrum(x2,Fs, 'spectrogram')
colormap(turbo)

.

[Michael Sugiarto]

Thankyou so much I will try it now!!

[Star Strider]

As always, my pleasure!