Detetcting time difference between peaks in 3D-Spectrogramm

1 次查看（过去 30 天）

显示更早的评论

Denny Muttathil 2017-1-31

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/322678-detetcting-time-difference-between-peaks-in-3d-spectrogramm

关闭： MATLAB Answer Bot 2021-8-20

在 MATLAB Online 中打开

I did a STFT on my audio signal. Then i noticed, i have very high amlitudes in a range between 1-1000hz. So i defined 3 frequency bands. Afterwards, i made waterfall-diagram by using the function "waterfall". I don't want to explain too much, since it would take so long. I recommend you, to try this code yourself. Just take any small wav-file, you can find and let it run on the computer. If you have questions concerning, my code please ask me in the comments. My wav-file has a duration from 16 seconds. At the beginning, i have very high peaks (amplitudes, i believe so). Is it possible to measure the time distance between those peaks and how can I write or do that?

%%MATLAB
%_________________________________________
[y,fs]=audioread('Undertale - Megalovania.wav');
% audioread = Read WAV-file
% y = Vector, which contains audio signal 
% fs = Sample Rate
% 'dontstopmenow' = WAV-file
%_________________________________________
%PARAMETER FOR STFT
%_________________________________________ 
t_seg=0.09; % Length of segment in ms
fftlen = 4096; %FFT-Points
%Defining the length of my frequency bands
f_LOW= 1:200;    % contain lower frequencies
f_MEDIUM= 201:600;  % contain medium frequencies
f_HIGH= 601:1000; % contain higher frequencies
%_______________________________________________________
segl =floor(t_seg*fs); 
% Length of segment, on which we use the fft
% "floor" rounds off the result
windowshift=round(segl/2); 
% size of window which goes to the next segment
window=hann(segl); 
%hann function
window=window.'; 
% From a row vector to a column vector
si=1; 
%Start index
ei=segl; 
%End index
N=floor( length(y)/windowshift - 1);
% Number of time segements in audio signal
f1=figure;
    % New window
      f=0:1:fftlen-1;
      f=f/fftlen*fs;
      % frequency vector
      Ya=zeros(1,fftlen);
      Yb = NaN(N, fftlen/2);
    %Plotting time segments!         
  for m= 1:1:N 
      y_a = y(si:ei);
      y_a= y_a.*window;
      Ya=fft(y_a, fftlen);
      Ya=abs(Ya(1:end/2));  
        %One-sided-spectrum
      Yb(m,:) = Ya;
        drawnow; %Updates graphical objects
      %figure(f1);
      %%L,M,H - Bands
      y_low = Ya(f_LOW);
      y_medium = Ya(f_MEDIUM);
      y_high = Ya(f_HIGH);
      si=si+windowshift; 
      % start index updated    
      ei=ei+windowshift; 
      % end index updated
      figure(f1);
      waterfall(f(f_LOW), (1:N)*windowshift/fs, Yb(:,f_LOW));
      xlabel('Frequency (Hz)'); 
      ylabel('Time (s)');
      set(gca,'YLim',[0 16])
      set(gca,'YTick',0:2:16) 
end