Models #1096
Shashuface
started this conversation in
General
Models
#1096
Replies: 2 comments
-
I've just started using this. the best I've found so far, and I think it's really good is an ensemble of MDX-NET Kim Vocals1, MDX-Net: UVR-MDX Inst HQ 3. MDX-Net: UVR-MDX Inst Main and Demucs v4 htdemucs_ft. Maybe all the loud music over the years but I don't hear any noise or artifacts of what was removed. |
Beta Was this translation helpful? Give feedback.
0 replies
-
Here is the model I been using, It has worked well for me, so maybe give it a try. Note: I updated UVR5 today, Feb 08, 2024 and it still worked great. |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I am working on making karaoke tracks. What are the best
models to use in UVR5, that would remove the main vocals, produce the best
instrumentals and keep the background vocals? I have tried different ensemble
combinations, but I am still hearing a lot of noise.
Beta Was this translation helpful? Give feedback.
All reactions