- Joined
- Apr 11, 2010
- Messages
- 955
On mvsep.com
I haven't tried it yet.
Also there is chatter about new general separation models coming that score higher than anything else publicly available. I guess we'll find out over time. The chatter was about "sami-bytedance" on the leader board here: h**ps://mvsep.com/quality_checker/multisong_leaderboard? but I don't know what the top 3 models are that have MUCH higher scores than even sami-bytdance".
I guess sami = SAMI (Speech, Audio & Music Intelligence) team at ByteDance AI Research lab
We have added the DrumSep model. This model produces a detailed separation of the drum track into 4 types: 'kick', 'snare', 'cymbals', 'toms'. The DrumSep model from this github repository is used. The model has two operating modes. The first (default) in the begining applies the Demucs4 HT model to the track, which extracts only the drum part. Next, the DrumSep model is applied. If your track consists only of drums, then it makes sense to use the second mode, where the DrumSep model is applied directly to the loaded audio. Demos available here.
I haven't tried it yet.
Also there is chatter about new general separation models coming that score higher than anything else publicly available. I guess we'll find out over time. The chatter was about "sami-bytedance" on the leader board here: h**ps://mvsep.com/quality_checker/multisong_leaderboard? but I don't know what the top 3 models are that have MUCH higher scores than even sami-bytdance".
I guess sami = SAMI (Speech, Audio & Music Intelligence) team at ByteDance AI Research lab