search

drethage / speech-denoising-wavenet

A neural network for end-to-end speech denoising
MIT License
673 stars 165 forks source link

Why does the trained model need clean audio files for inferencing on the same noisy audio files? #10

Closed DillipKS closed 6 years ago

DillipKS commented 6 years ago

Ideally, while inferencing the trained model should take the noisy audio data and output the denoised/cleaned version of the same audio files. I have some noisy audio files on which I want to run the pretrained model and get the clean files. But, while inferencing, it asks for path to clean audio data too which I don't have and the model doesn't run without it. Why is a previously trained model asking for clean audio files? @drethage

jordipons commented 6 years ago

If you remove the "--clean_input_path data/NSDTSEA/clean_testset_wav" part of your script call, it should work.

DillipKS commented 6 years ago

Thanks for the info @jordipons . Can you also clarify if the training/test data needs to be in .wav format or it can support any other format? I have audio data in .mp3 format and converting it to .wav is resource intensive. How can I train and test the wavenet model with data in mp3 format itself?

drethage commented 6 years ago

During inference, the clean audio files are only used for computing statistics on the output.