How I get the frequencies in Hz of a specific signal?
3 views (last 30 days)
Show older comments
I created a matlab code that gives me in output a signal that starting at 80 Hz go up till 800Hz with the following code
% % This code will generate a saltatory stimulus of 10 Hz up to 800 Hrz (the stimulus was frequency filtered in audacity to exclude frequencies bellow 80 and above 1000Hz)
duration = 80;
freq_step = 10;
sampling_rate = 44100;
% Generate the time samples
t = linspace(0, duration, sampling_rate * duration);
%generation of the audiowave with the increasing frequency steps
freq_range = freq_step * (0:length(t)-1) / sampling_rate;
audio_waveform = 0.5*sin(pi * freq_range.* t);
audio_waveform(1:(8*sampling_rate)) = [];
%normalization of the audiowave form
%audio_waveform = audio_waveform / max(abs(audio_waveform)); %this
%commented
%eliminate the normalization of the wave form
%create the audioplayer object
player = audioplayer(audio_waveform, sampling_rate);
%play(player);
%save the file as WAV file format
audiowrite("output_sound1.wav", audio_waveform, sampling_rate);
Now my real gole is to given a signal as an input I would like to obtain the frequencies that my audio has allong the whole duration..but i dont know how to get the frequencies ... I thought that by using a STFT i will get it and i used my signal (which i know should have frequencies from 80 upt till 800 with an increas of 10Hz in every step) but I cannot get the array and spectrogram or something else that should tell me the frequency components of my signal allong the duration something like 1s = frequency 80 Hz, 2s = frequency 90 Hz and so on till 80s = frequency 800Hz. what is wrong with my code?
filename = 'output_sound1.wav'; % Replace wih the actual file name
[audio_data, sample_rate] = audioread(filename);
% now i need to calculate the number of samples in mz audio data
num_samples = length(audio_data);
%calculate the frequencies in the file this with the FFT
frequencies = (0:num_samples-1)* sample_rate / num_samples;
% Initialize arrays to store dB and volt values
amplitude_dB = zeros(length(frequencies), 1);
amplitude_volt = zeros(length(frequencies), 1);
for i = 1:length(frequencies)
freq_index = (i);
% Calculate amplitude in dB and volts78
amplitude_dB(i) = 20 * log10(abs(audio_data(freq_index)));
amplitude_volt(i) = abs(audio_data(freq_index));
end
% Create a table
Frequency_Hz = frequencies';
frequency_table = table(Frequency_Hz, amplitude_dB, amplitude_volt);
% lets try doing the spectrogram of my signal
y_stft = stft(audio_data);
N = length(audio_data);
y = audio_data-mean(audio_data);
fr= sample_rate * (0:floor((1/2)/N));
s = abs(y_stft(1:floor(N/2+1)));
%plt the single-sided magnitide spectrum of the signal
% figure
plot(fr,s);
title("single sided spectrum of my signal");
xlabel("frequenzy");
ylabel("SFTF of my signal");
window = 100;
overlap = window/2;
nfft= 1024;
[S, f, t] = spectrogram (audio_data, window, overlap, nfft, N);
%
% %conversion of the spectrogram in dB air scale
%
S_dB = 10*log10(abs(10));
% %plot the spectrogram
%
figure
imagesc (t, f, S_dB);
axis xy;
xlabel("Time (s)");
ylabel("Frequency(Hz)");
title("Spectrogram of my generated signal");
colorbar;
0 Comments
Answers (1)
Les Beckham
on 16 Aug 2023
I think this is close to what you are looking for.
% % This code will generate a saltatory stimulus of 10 Hz up to 800 Hrz (the stimulus was frequency filtered in audacity to exclude frequencies bellow 80 and above 1000Hz)
duration = 80;
freq_step = 10;
sampling_rate = 44100;
% Generate the time samples
t = linspace(0, duration, sampling_rate * duration);
%generation of the audiowave with the increasing frequency steps
freq_range = freq_step * (0:length(t)-1) / sampling_rate;
audio_waveform = 0.5*sin(pi * freq_range.* t);
audio_waveform(1:(8*sampling_rate)) = [];
%normalization of the audiowave form
%audio_waveform = audio_waveform / max(abs(audio_waveform)); %this
%commented
%eliminate the normalization of the wave form
%create the audioplayer object
% player = audioplayer(audio_waveform, sampling_rate);
%play(player);
%save the file as WAV file format
% audiowrite("output_sound1.wav", audio_waveform, sampling_rate);
m = islocalmax(audio_waveform); % find indexes of the signal peaks
t_plot = t; % make a copy of the time vector to use for plotting
t_plot(1:(8*sampling_rate)) = []; % truncate the time vector the same way the audio_waveform was truncated
f = 1 ./ diff(t_plot(m)); % frequency is inverse of time between signal peaks
t_plot = t_plot(m); % extract the time at each of the peaks
% whos
plot(t_plot(2:end), f); % skip first time point since diff reduces vector length by one
grid on
xlabel('Time (sec)')
ylabel('Frequency (Hz)')
0 Comments
See Also
Categories
Find more on Spectral Measurements in Help Center and File Exchange
Products
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!