search

tsurumeso / vocal-remover

Vocal Remover using Deep Neural Networks
MIT License
1.47k stars 215 forks source link

How to use the Karaoke model #169

Open mc261670164 opened 6 months ago

mc261670164 commented 6 months ago

Thanks for your great open source work! I had seen the Karaoke model used in UVR, and I wonder how can use the Karaoke model in this project. Thanks very much!

mc261670164 commented 6 months ago

Yes, I see in UVR that Vocal-Remover module can be used with Karaoke model. So I wonder how to use that model in this project. Can you give me some guidance? Thank you!

At 2023-12-26 23:37:36, "Goldp-js" @.***> wrote:

Do you mean using the UVR models in this project?

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

Goldp-js commented 6 months ago

I had already tried using the UVR models in this project, since they have different architectures but the same .pth extension. I tried to find a way to adapt them, but a problem arose: the dimensions of the VR models are not compatible with this project.

Goldp-js commented 6 months ago

The model for this project is compatible with UVR, but the UVR models are not compatible with this project. Unless they release a command-line version of UVR, that is.

aufr33 commented 6 months ago

Hi. I was creating this model for the UVR (VR fork). Since it uses modified tsurumeso code, it will only work correctly in UVR.

MinzChan commented 6 months ago

Hi. I was creating this model for the UVR (VR fork). Since it uses modified tsurumeso code, it will only work correctly in UVR.

!python inference.py --gpu 0 --pretrained_model models/HP2-4BAND-3090_4band_2.pth --input './input/test.wav' not use vocal-remover baseline.pt but ultimatevocalremovergui HP2-4BAND-3090_4band_2.pth Your VR fork project doesn't seem to work properly