Scaled frequency magnitude spectrum using FFT

Could someone please check and advise if the MatLab code shown below is correct to plot the scaled time waveform and the scaled frequency magnitude spectrum? I wish to get a sharp spike, remove the side-lobes, and get something as shown below.

close all;

clear all;

clc;

audioread('DTMFDigit3.wav');

info = audioinfo('DTMFDigit3.wav');

display(info);

[Dialtone, fs] = audioread('DTMFDigit3.wav');

sound(Dialtone, fs)

N = length(Dialtone);

idx = 0:N-1;

t = idx*(1/fs);

figure(1);

plot(t, Dialtone)

title('Dial Tone Time Waveform'); % plot title3

xlabel('Time'), ylabel('Amplitude'); % plot axis labels

N= 2^nextpow2(N);

k=(-N/2):1:(N/2)-1;

f=(fs/N)*k;

FFT_dialtone = fft(Dialtone,N);

FFT_dialtone = abs(fftshift(FFT_dialtone))/N;

figure(2);

plot(f,FFT_dialtone);

xlabel('Frequecy -fs/2 to fs/2');

ylabel('Magnitude');

title('Spectrume of dialtone');

Answer 1

[Rahul]

As per the audio file "DTMFDigit3.wav", it seems that you are trying to extract the frequency information of DTMF digit 3 dialtone. As per the DTMF frquency table given in wikipedia page, the frequencies associated with it are 697 Hz and 1477 Hz. The plot that you have provided is for DTMF digit 0 dialtone. Nevertheless, you can use following code to extract the frequency information of any signal.

close all;
clearvars;
clc;

%% Read the audio file
info = audioinfo('DTMFDigit3.wav');
display(info);
[Dialtone, fs] = audioread('DTMFDigit3.wav');
sound(Dialtone, fs) % play the audio file 
N = length(Dialtone); % extract the length of the signal
t = (0:N-1)*(1/fs); % time period of the signal 
figure(1);
plot(t, Dialtone)
title('Dial Tone Time Waveform');

%% Extract the frequency information of the dialtone using FFT
Y = fft(Dialtone);
% Compute the two-sided spectrum P2. 
% Then compute the single-sided spectrum P1 based 
% on P2 and the even-valued signal length L.
P2 = abs(Y/N);
P1 = P2(1:N/2+1);
P1(2:end-1) = 2*P1(2:end-1);

%% Plot the frequency spectrum
f = fs*(0:(N/2))/N;
figure(2)
plot(f,P1) 
title("Single-Sided Amplitude Spectrum of X(t)")
xlabel("f (Hz)")
ylabel("|P1(f)|")

%% remove the unwanted sidelobes using thresholding
thresh_val = 0.02;
P1_smooth = zeros(1, length(f));
for ii = 1:length(f)
    if P1(ii)>thresh_val
        P1_smooth(ii) = P1(ii);
    end
end

figure(3)
plot(f, P1_smooth)
title("Single-Sided Smoothed Amplitude Spectrum of DTMF Digit 3")
xlabel("f (Hz)")
ylabel("|P1 smooth|")