open-unmix-pytorch

For MIDI: https://github.com/Music-and-Culture-Technology-Lab/omnizart and https://basicpitch.spotify.com/ They work better if you apply some source separation before (e.g, https://github.com/sigsep/open-unmix-pytorch, https://github.com/facebookresearch/demucs, or https://mvsep.com) Still, I think the best results are from proprietary models (specifically https://www.ableton.com/en/manual/converting-audio-to-midi/ and https://www.celemony.com/en/melodyne/what-is-melodyne)

In my first post, quite a lot of alternatives were discussed: https://news.ycombinator.com/item?id=36707877 The model I'm using is called Open-Unmix (https://github.com/sigsep/open-unmix-pytorch). In 2021, there was an update to Open-Unmix to include new weights, UMX-L, which made it perform better than it used to on the older weights (UMXHQ). In the grand landscape of music demixing, I don't think UMX-L is near the top anymore. _However_, the demixing performance of freemusicdemixer.com is very close to the full PyTorch performance of Open-Unmix UMX-L, despite the tricks I needed to get it working in the browser, such as splitting up the inference to operate on segments of the song, or making the LSTM operate on streaming segments rather than holding the entire track in the LSTM memory. In my first release, I loaded and did inference on

In the free-music-demixer repo, there is a `file_demixer` utility [1] which you can compile using the CMakeLists file - it will use the same quantized weights and produce the same "lower quality" output as the website. Clone the git repo with submodules because it compiles the vendored libnyquist library to load audio files. Of course, you can always run the upstream umxl model easily: ``` pip install openunmix # pulls in pytorch ``` Installing openunmix installs the `umx` cli: ``` $ umx --help usage: umx [-h] [--model MODEL] [--targets TARGETS [TARGETS ...]] [--outdir OUTDIR] [--ext EXT] [--start START] [--duration DURATION] [--no-cuda] [--audio-backend AUDIO_BACKEND] [--niter NITER] [--wiener-win-len WIENER_WIN_LEN] [--residual RESIDUAL] [--aggregate AGGREGATE] [--filterbank FILTERBANK] [--verbose]

There's maybe something useful for you here? https://github.com/facebookresearch/demucs https://github.com/sigsep/open-unmix-pytorch https://github.com/bytedance/music_source_separation https://github.com/deezer/spleeter

Spleeter and UMX are both FOSS [0] [1], whereas, as far as I can tell, Cassiopeia is not. [0] https://github.com/deezer/spleeter/blob/master/LICENSE [1] https://github.com/sigsep/open-unmix-pytorch

I'm trying to estimate what note is being played in an instrumental track. I'd tried to approach this problem with by splitting the track into its different instruments with open_unmix [0], and then using librosa pitch tracking [1], and then cleaned the results, but unfortunately, the midi came out totally unrecognizable. I thought about approaching this problem again when I saw the recent post on audio source separation [2], but that still only solves the problem of separating the vocals from the instrumental track and into separate instruments. Is there a better/more state of the art method to pitch estimation than librosa's piptrack, or am I just using piptrack improperly? [0] https://github.com/sigsep/open-unmix-pytorch [1] http://man.hubwiz.com/docset/LibROSA.docset/Contents/Resources/Documents/generated…

Another recent open source contender for source separation is Open Unmix: https://github.com/sigsep/open-unmix-pytorch/ I’ve not had time to try it yet but have read good things.

I wonder how this compares to Open Unmix (https://github.com/sigsep/open-unmix-pytorch), that one calls itself state-of-the-art as well and is done in collaboration with Sony from what I see of their paper.