Pro site released
·
89 commits
to main
since this release
I've been working on new models after the conclusion of the Sound Demixing Challenge 2023
Ensemble models dominated, with mvsep using an interesting technique of inference augmentation with the inverted waveform:
stems = model_1(waveform)
inverse_stems = model_1(waveform * -1.0) * -1.0
I'm launching the pro site with special performing ensembles of demucs:
