Reproduce “A Tutorial on Generative Models for Music Audio” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Mateusz Modrzejewski, Joanna Kurkowska, Dorien Herremans

The authors present a tutorial and field guide to music generation that links tasks to representations, generation methods, conditioning routes, and samplers. It explains how these choices shape systems for coordinated songs, editing, and real-time use. A controlled comparison finds autoregressive and flow-matching methods trade off quality, speed, and controllability, without a dominant winner; broader system tables cannot isolate causes.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.