muzic VS abjad

We open our code and checkpoint here: muzic/getmusic at main · microsoft/muzic (github.com)

Code: https://github.com/microsoft/muzic

There's a lot of research right now into music generation (example) that could potentially lead to songs completely written, performed, and produced by software. While I don't think those will be topping the charts, the same tech could definitely be used by humans to generate and expand on ideas, and the success of things like Twitter bots powered by GPT-3 makes me think there could be a niche for botpop. On the extreme end, imagine a service like Pandora, except it generates brand new songs tailored to your preferences.

Muzic (https://github.com/microsoft/muzic) is a research project on AI music that empowers music understanding and generation with deep learning and artificial intelligence. Muzic was started by some researchers from Microsoft Research Asia.

We initially release the code of 5 research work: MusicBERT, PDAugment, DeepRapper, SongMASS, and TeleMelody. We will release the code of more research work in the future.

And for the intrepid folks who prefer algorithmic or computer-assisted composition, there's Abjad, a Python-based wrapper around LilyPond.

https://abjad.github.io/

The abjad library looks really interesting, as it basically gives a Python API for Lilypond. It would help me do all the main features I want. However, from the way it does its Lilypond objects, each song would have to output its own PDF. I don't see how I can use something like PyLaTeX to concatenate all those existing PDFs and make nice hyperlinks and indexes for it all.

This could be done in many other ways. Indeed this is not even the standard way (that would be coding a 2to3 fixer). The practical solution is not general (for instance, it does not support nested classes). Also, it has problems with comments (as we lose them in the AST). But, for the practical purpose (patching abjad) it was good enough (You can see the result here.