Any plans to push the source code? Really looking forward to it.

Question

Any plans to push the source code? Really looking forward to it.

RahulBhalley opened this issue 3 years ago · 7 comments

Answer 1 · 2021-11-26T23:21:01.000Z

I am currently implementing this. 😅 I will take some time for implementation and experiments.

Answer 2 · 2021-11-27T04:09:47.000Z

Sure, no issues. :) It would really help me also implement it on my end if you could please share some references repositories that you're probably looking at right now to implement NANSY.

Regards
Rahul Bhalley

Answer 3 · 2021-11-27T06:41:28.000Z

Hmm, I'm currently trying to implement it from the paper. librosa/torchaudio/parselmouth was helpful.

Answer 4 · 2021-11-27T19:50:42.000Z

Ok, thanks!

Answer 5 · 2022-03-31T14:56:02.000Z

Hi @rosinality any updates on this?

Answer 6 · 2022-03-31T16:02:14.000Z

I have almost implemented the model, but I found aligning features which have different sampling ratios/window sizes (wav2vec features vs mel, yingram features) is not straightforward. Currently I am stopped at that point.

Answer 7 · 2022-03-31T16:24:32.000Z

Ok, thanks for updating me. :)