Speech enhancement

[AlexandderGorodetski]

Hi everyone,

I have some very noisy audio data with a lot of added noise, distortion from saturation, and bit quantization errors.

I’d prefer not to spend time training a new K2 model with these types of augmented errors.

Can you recommend a speech enhancement model that I can use on this data to improve the word error rate (WER)?

Thanks a lot, AlexG.