search

fjiang9 / NKF-AEC

Acoustic Echo Cancellation with Nerual Kalman Filtering
240 stars 60 forks source link

Some problems on real sequences. #6

Closed c8x1 closed 2 years ago

c8x1 commented 2 years ago

I tried to reproduce your results, but I found that there are two phenomena (bad results) in some real sequences, both of which also appear in the model you provide.

  1. may be that the filter diverges (the estimated echo energy is unusually high);
  2. the estimated echo have a howling-like sound (which doesnt exist in the reference signal); Most of these two happens at the beginning of the sentence(but not necessarily at the beginning of the sequence).

Did you find anything like this?

fjiang9 commented 2 years ago

@c8x1 Could you provide the test audio samples you used?

c8x1 commented 2 years ago

sry i cant provide them because of privacy reasons. its just some real captured audio. But the audio itself is not special. I blame it on the lack of stability of the filter, maybe some post-processing.