I am trying to write a program that would help me match some small sound. (For e.g. snapping of two fingers in an audio clip). I took a sample with had the desired sound in it, and plotted the frequency distribution (after doing fourier transform on the audio input). I ended up with a graph as follows: enter image description here

I would want to know what if the snap sound itself got less louder...would the amplitude in the graph above scale down but maintain the general shape?

Secondly, what does it mean if such a distribution "shifts", say, to the right, but maintains the same "shape"? In this case, can we conclude that it is the same sound?

Ps: although I say "shape" they look erratic yet they have a pattern to it. This will be more visible in the animation I shall post in a short while.... https://youtu.be/bMYYZ6eGgsc


1 Answer 1


The audio file consists of the snap sound added to some background noise (which will always be there, and probably dominates at both low and high frequencies). If the snap got softer, then the background noise would be more important, and the Fourier transform would begin to look more like the background noise. The particular shape of the background noise depends on the acoustics of the room you're recording in, the qualities of your microphone, the idiosyncracies of your audio processing software, and several other factors. All of these things are hard to predict, so if your sound is too soft, there's really not much you can say. However, if your snapping sound is significantly louder than the background, then yes, basically the amplitude of the Fourier transform will just decrease (if you restrict yourself to the frequency interval in which the snapping sound actually is significantly louder than the background).

A shift to the right would indicate that your sound is becoming higher-pitched. As such, I wouldn't call it the same sound.

  • $\begingroup$ Thanks for your answer. Would the resulting sound produced by the slight shift appear more or less similar or same to the human ear? $\endgroup$
    – deostroll
    Apr 19, 2018 at 4:18
  • $\begingroup$ The human ear would definitely be able to tell the difference; we're pretty good at distinguishing different pitches, as that's the foundation for basically all music. $\endgroup$ Apr 19, 2018 at 4:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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