Audio is a sequence of waveforms which have peaks, troughs, 0, just like below. audio_waveform

How to define the x axis (y = 0) in the picture, how to choose the 0 point in the waveform since there are so many points on the waveform?

Why not move the waveform above or below the x axis(y=0)?

This question seems a little stupid. But I still hope someone can help with it.


1 Answer 1


You can't produce a DC component with speech, so the line $y=0$ is defined by the speech signal having zero mean.

  • $\begingroup$ You are right Matt, but DC offset can be very often added by recording device itself. $\endgroup$
    – jojek
    Aug 12, 2014 at 8:07
  • 1
    $\begingroup$ @jojek: That's true but that would be an artefact and the DC component does not carry any information about the speech signal. $\endgroup$
    – Matt L.
    Aug 12, 2014 at 8:31
  • $\begingroup$ Could I use median instead? What consequences would it have? $\endgroup$
    – hans
    Apr 12, 2021 at 23:55

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.