xiph/rnnoise

How to prepare training data

jacky-ge opened this issue · 0 comments

The signal.raw file should be clean speaker's voice, but I don't know the noise.raw file is pure noise or the result of adding noise to the signal.raw file