2
$\begingroup$

I've got this unusual spectrogram and I am really curious what these circles could mean. Has anyone ever seen anything like this? What could be the source of the signal?

spectrogram

$\endgroup$
  • $\begingroup$ Would it be possible to provide some more information about this? For example, a plot of the signal in the time domain, the package, function name and parameters you are passing to it and whether you built the spectrogram function? This has an "audio explanation" but its boring and unlikely and the pattern is suspiciously log and regular. Could be a bug in overlap-(add/save) or poor choice of parameters. $\endgroup$ – A_A Jan 4 '16 at 9:33
  • $\begingroup$ That's what I'm looking for, i.e. what kind of signal could produce this spectrogram! First, I thought that the pattern is due to a 'bug' or improper usage. But while the circles are regular they are not regular enough... Do you mind to share your "audio explanation" or just give me some clue? $\endgroup$ – zer0hedge Jan 4 '16 at 10:33
  • $\begingroup$ OK, straightforward answer from what is provided: If this is an audio signal, then it could have resulted from something that sounds like this freesound.org/people/Timbre/sounds/122950 until the pause followed by its reversed version. But your case doesn't contain this much noise and seems to also have some echo added to it because faint "circles" appear shifted in time too. But without a bit more information it is impossible to provide any more meaningful help (i.e. whether this could be the result of bug or erroneous parameters). $\endgroup$ – A_A Jan 4 '16 at 12:14
  • 2
    $\begingroup$ It looks much like some Moiré pattern $\endgroup$ – Olli Niemitalo Jan 4 '16 at 13:52
2
$\begingroup$

In Understanding FFT Overlap Processing, p. 10, it is suggested that uniform of constant windows, for certain hop patterns, with repeated pulses, could generate such artifacts.

Spectrogram with Moiré

$\endgroup$
2
$\begingroup$

I'm not able to comment yet but could it be some kind of FM Synthesis? Like Olli Niemitalo wrote, it looks like Moiré pattern. I found a spectrogram which looks a little bit similar: enter image description here

$\endgroup$

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.