I have 3 wav files:

  1. The original recording of a viola note
  2. A noisy re-recording of the original clip
  3. A different noisy re-recording of the original clip

I'm looking for a way to show which of the two noisy re-recordings is better than the other, and/or more similar to the original.

I can plot the time-domain waveforms and spectrograms which show that the signals are vaguely similar, but ideally I could find a measurement that gives me a single number - something like "recording 1 = 35% quality, recording 2 = 63% quality". I've tried "corcoeff".

In the image below, the original is in blue, the decent recording is in green, and the bad one is in red. enter image description here

  • 2
    $\begingroup$ Does this answer your question? Audio Quality Assessment $\endgroup$ – MBaz Mar 4 '20 at 0:52
  • $\begingroup$ Maybe you could compare SNRs of the audio signals $\endgroup$ – DSP Novice Mar 5 '20 at 4:31
  • $\begingroup$ There are no actual answers in the above question. $\endgroup$ – Sevag Mar 5 '20 at 15:01

Your Answer

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

Browse other questions tagged or ask your own question.