I ran FFT on three audio files and found that the results for some have more peaks than the other. Could anyone give me any conceptual explanation as to what determines these peaks? Below are plots of the audio sound envelope (left) and the corresponding FFT results (right). If I look at the plots on the right, e.g., the FFT for audio 1 has no peaks between 0.33 and 0.67 but the FFT for audio 2 and 3 do. I am not quite sure whether these additional peaks are meaningful or just artifacts of my particular sound file and FFT setup. More explanations about the audio files below.

enter image description here

Each peak in the audio plots represents a sinewave tone. The smallest inter-tone interval is 187.5 ms. Other inter-tone intervals could be 2x, 4x, or 6x187.5 ms. Each audio file lasts 5x3 seconds. The first row is the result from the exact same 3-second rhythm looped 5 times. The last two are the repetition of different 3-s rhythms. The inter-tone intervals in the three files are as follows:

enter image description here

Here's my FFT script:

n = length(sound);
nyquist = srate/2; 
f = linspace(0, nyquist, floor(n/2)+1); 
fourier_coef = fft(snd);
fourier_coef = fourier_coef/n;
amp = abs(fourier_coef);
amp = amp .*2;

Thank you.



Peaks are determined by energy in a given frequency range which can be more or less visible depending on the dimension of frequency bins.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.