I'm looking for your opinions.

I have obtained a Speech Signal and computed a STFT on this signal. I have split the signal into blocks of (100 x 256) and computed the STFT based on different Windowing functions. But, they all bring back similar results (outputted on the same block) I know this might make sense, but, I assumed that because I was using different Windowing functions, it would have a major impact on the results..


Hamming ^^ Note how the block contains no actual speech, just white noise as it's the first block. Does this look normal, or, is there something wrong here?

  • $\begingroup$ Related: dsp.stackexchange.com/questions/1618/… and dsp.stackexchange.com/questions/208/… $\endgroup$
    – lmjohns3
    Commented Aug 17, 2013 at 15:47
  • 2
    $\begingroup$ What have you plotted? Where are the axis names? How many points of FFT did you use? Did you zero pad? $\endgroup$
    – user13107
    Commented Aug 17, 2013 at 16:04
  • 1
    $\begingroup$ You might also want to show plots of actual speech data, since it sounds like that's what you're planning to apply your window function + STFT to. Plotting transforms of noise isn't always very clarifying. $\endgroup$
    – lmjohns3
    Commented Aug 17, 2013 at 16:07

1 Answer 1


This is a rather general answer : I would say that, although each window function has specific characteristics that make it more or less suitable for specific situations, in this context you're not going to see a huge difference in the STFT of a windowed speech signal.

In particular, the two windows that you're comparing are quite similar, so I'd expect that you'd get pretty similar STFT results. You might try two windows that are quite different, like a rectangular and a Hamming or something like that, to see just how much difference those make in your STFT, and then use that difference as a sort of upper bound on the difference that you're likely to see from other windows.


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.