0
$\begingroup$

I have one audio file containing human speeches, a lot of ambient noises like audience laughter, birds chirping, sounds of natures etc..

Now I want to separate it into two audio files where one should contain all human vocals and other should contain all ambient noises like audience laughter, birds chirping, sounds of natures etc..

I have look into some methods like voicefilter paper from google AI but the result here is the human speech and it's not possible to subtract the resultant with the source sound to get the audio file for other ambient noises, since we can't just assume they are a linear combination.

$\endgroup$
  • $\begingroup$ Did you look at any source separation methods? $\endgroup$ – havakok Dec 5 '19 at 9:00
2
$\begingroup$

What you are trying to archive is very hard if at all possible. The pinnacle of research today (including my own) is striving for good results on such tasks. DCASE challanges engage with similar tasks with some interesting results, though the make assumption which cannot be generalized to your case. In DCASE2019 task 3, for example, they assume up to 2 timewise parallel sources where in your case there are no such limitations.

With respect to your conditions and constraints, I would advise looking into source separation and\or classification. If your background audio has a repetitive nature, this paper may help.

$\endgroup$

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.