search

nomadkaraoke / python-audio-separator

Easy to use stem (e.g. instrumental/vocals) separation from CLI or as a python package, using a variety of amazing pre-trained models (primarily from UVR)
MIT License
514 stars 86 forks source link

[Bug]: "Unsupported Model File" error for HTDemucs_6s #161

Closed IneffableBunch closed 1 day ago

IneffableBunch commented 2 days ago

Describe the bug

audio-separator --list returns: "Demucs v4: htdemucs_6s": { "5c90dfd2-34c22ccb.th": "https://dl.fbaipublicfiles.com/demucs/hybrid_transformer/5c90dfd2-34c22ccb.th", "htdemucs_6s.yaml": "https://github.com/TRvlvr/model_repo/releases/download/all_public_uvr_models/htdemucs_6s.yaml" }

but loading 5c90dfd2-34c22ccb.th does not work.

Have you searched for existing issues? 🔎

Screenshots or Videos

No response

Logs

No response

System Info

Operating System: Windows 
Python version: 3.10
Other...

Additional Information

No response

beveradb commented 2 days ago

That's just a little inconsistency in the model list output because there are multiple sub-files for the demucs models (which could probably be tidied up in the -l output)

to use that demucs model use the model filename "htdemucs_6s.yaml" instead 😄

Lixi20 commented 2 days ago

like this: separator = Separator() separator.load_model(model_filename='htdemucs_6s.yaml')