search

haoheliu / voicefixer_main

General Speech Restoration
https://haoheliu.github.io/demopage-voicefixer/
MIT License
272 stars 52 forks source link

Train with 16k data #9

Open victorVoice opened 2 years ago

victorVoice commented 2 years ago

Hi, thanks for the awesome work. I have a question here, if i train the model using 16k data , will the trained model guarantee the performance?

haoheliu commented 2 years ago

Modeling 16k data is usually easier than 44.1k data because of the smaller dimension. Personally, I think the 16k model would be easier to train and get a better result.

victorVoice commented 2 years ago

Thanks for the quick reply, i will do some experiments on the 16k data

pandysong commented 2 years ago

@ws-choi Interesting. I am thinking of 16KHz sampling rate as well for realtime conference. It would be appreciating if you will get a good result.