MusGO Framework: Assessing Openness in Music-Generative AI

This website builds on the paper MusGO: A Community-Driven Framework for Assessing Openness in Music-Generative AI, authored by Roser Batlle-Roca, Laura Ibáñez-Martínez, Xavier Serra, Emilia Gómez, and Martín Rocamora, and published in the Proceedings of the 26th International Society for Music Information Retrieval Conference (ISMIR 2025). It serves not only as a companion to the publication, but also as a living resource, which is continuously updated and shaped by contributions from the community.

VampNet

Northwestern University and Descript Inc · 2023

A masked acoustic token modeling approach to music generation.

Application types: Audio-to-audio

Architecture:Non-autoregressive transformer

Essential categories

Category	Status	Notes
Source code	✔︎ Open	System source code is available and accessible.
Training data	✘ Closed	Dataset is not available nor is it completely described.
Model weights	✔︎ Open	Model checkpoints are available. The weights of the models are licensed https://creativecommons.org/licenses/by-nc-sa/4.0/deed.ml
Code documentation	✔︎ Open	Codebase is well documented.
Training procedure	✔︎ Open	Described in the article, with multiple details on architecture, hyperparameters and GPU usage.
Evaluation procedure	~ Partial
Research paper	✔︎ Open	Accepted at ISMIR 2023.
Licensing	✔︎ Open	MIT License.

Desirable categories

Category	Status	Notes
Model card	∅ Not included
Datasheet	∅ Not included
Package	∅ Not included
User-oriented application	⭐ Included	Gradio UI is provided in the codebase.
Supplementary material page	⭐ Included	Complementary demo page with sound examples and demonstration of the model's capabilities is available.

Raw YAML file with complete evaluation.