search

haoheliu / voicefixer_main

General Speech Restoration
https://haoheliu.github.io/demopage-voicefixer/
MIT License
274 stars 54 forks source link

Is it possible for realtime processing? #11

Open pandysong opened 2 years ago

pandysong commented 2 years ago

As titled

haoheliu commented 2 years ago

@pandysong Sorry for the late reply. The model is non-autoregressive and mainly CNN-based. So if running on a GPU with some optimizations, it could be functioning in real-time.

pandysong commented 2 years ago

@haoheliu Thanks. I am thinking of porting to an SoC with NPU available on it, if a pre-trained model would be available. Do you think it is possible that a pre-trained model will be online?

haoheliu commented 2 years ago

@pandysong Porting to an SoC, that's cool! Actually, the pre-trained model is available here. You can play with it if you are interested.

pandysong commented 2 years ago

@haoheliu Thanks. I think I looked at it before. and I will check it again. Thank you for your work.

SutirthaChakraborty commented 1 year ago

Did you implement in realtime ? @pandysong