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.

RAVE

IRCAM · 2021

Fast and high-quality audio waveform synthesizer.

Application types: Audio synthesisStyle transfer

Architecture:VAE

Essential categories

Category	Status	Notes
Source code	✔︎ Open	Source code is available, including data preparation, model architecture, training pipeline and inference.
Training data	✘ Closed	Training data is internal and just briefly described.
Model weights	✔︎ Open	Several pretrained models are provided.
Code documentation	✔︎ Open	Code is thoroughly documented, allowing for understanding and reproducibility of the model.
Training procedure	✔︎ Open	Training procedure is described.
Evaluation procedure	✔︎ Open	Evaluation procedure is described, and the implementations used for the evaluation metrics are referenced.
Research paper	~ Partial	Preprint only. No peer-reviewed version.
Licensing	~ Partial	System covered by a CC BY-NC 4.0 license.

Desirable categories

Category	Status	Notes
Model card	⭐ Included	Model card is available.
Datasheet	∅ Not included	Not available.
Package	⭐ Included	Pip package is available.
User-oriented application	⭐ Included	Real-time implementation of the model and tutorials are provided.
Supplementary material page	⭐ Included	Demo page is available, showcasing the model's capabilities and providing sound examples of its use.

Raw YAML file with complete evaluation.