search on Github....I'll try and post a link later.
I'm trying to install Anaconda and open unmix now..
The open unmix is mixing conda and pip commands. Doesn't seem like a good idea but maybe it's OK inside a conda environment? I guess I'm not clear on that part. For myself I haven't got umxl working here with GPU (although other python tools do). I can run the Docker version (CPU only) however.
The latest Open Unmix (umxl): GitHub - sigsep/open-unmix-pytorch: Open-Unmix - Music Source Separation for PyTorch
Demucs: GitHub - facebookresearch/demucs: Code for the paper Music Source Separation in the Waveform Domain
Spleeter GUI: SpleeterGUI | Maken It So
Ultimate Vocal Remover: GitHub - Anjok07/ultimatevocalremovergui: GUI for a Vocal Remover that uses Deep Neural Networks.
One of the problems with these is that the metric they use to compare tools and drive the technology forward I don't feel necessarily corresponds to "good" sound but just more separation.