White noise for 48kHz model

sp-uhh / sgmse

Score-based Generative Models (Diffusion Models) for Speech Enhancement and Dereverberation

MIT License

454 stars 69 forks source link

White noise for 48kHz model #50

Closed isomov closed 2 months ago

isomov commented 2 months ago

Hello, thanks for the great work and 48kHz checkpoint! I was trying to check it on my 48kHz audio, however it produces white noise. I did not change code and installed my environment using requirements_version.txt. This is how I run the model: "python enhancement.py --test_dir noisy48/ --enhanced_dir enhanced/ --ckpt EARS\ checkpoints/ears_wham.ckpt" Can you help me with that, please?

julius-richter commented 2 months ago

I just checked again, and it works on my side.

Have you pulled the latest commit?

Can you show me the data that you want to enhance?