I'm unable to compute $f_0$ (fundamental frequency) in librosa feature extraction. By ready much in github issues check the second comment. I see that $f_0$ is correlated with some other features (like zero_crosssing) easy to extract from librosa.

What are these features except for that one I give? Is this correlation super high so that they can replace $f_0$?

data sample

site for data

  • $\begingroup$ I mean fundamental frequency, I edited the post. $\endgroup$ – abdoulsn Nov 17 '19 at 18:11
  • $\begingroup$ updated again the post. $\endgroup$ – abdoulsn Nov 17 '19 at 18:18
  • $\begingroup$ so, are you specifically interested in a YIN-based fundamental frequency estimate, or is the question more generally, "what features correlate well with fundamental frequency"? In either case, are we looking at a specific kind of signal (speech, instrumental music, birdsong, cars...)? $\endgroup$ – Marcus Müller Nov 17 '19 at 18:29
  • $\begingroup$ This is generally question, I'm using speech emotion data(.wav audio file with one expression). $\endgroup$ – abdoulsn Nov 17 '19 at 18:30
  • 1
    $\begingroup$ Like, upload one of your .wav files somewhere and link to that! $\endgroup$ – Marcus Müller Nov 17 '19 at 18:35

Here is the answer as responded here in github.

| improve this answer | |

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.