multimodal-art-projection/YuE ? reverse-engineered prompt

Build me the YuE2 music generation project so I can try it end to end from a fresh clone.

I want to be able to install it, run the example generator, and produce a full song from a simple JSON request, then save the audio plus the score and other artifacts in an output folder. Please make sure the main Python entry points work for creating a new song, making a cover from an ABC score, and editing a score and rendering a new version. If anything depends on current model or setup instructions, look up the latest docs online and wire it up cleanly.

Keep it simple and easy to use, with clear examples I can run locally on a Linux machine with Python 3.12 and a CUDA GPU. If there are any missing pieces in the repo, please fill them in so the quick start works without me having to hunt around.