Diff-A-Riff

Sony Computer Science Laboratories Paris and Queen Mary University of London · 2024

Generative model for instrumental accompaniments for any musical audio context.

Application types: AccompanimentAudio-to-audioText-to-music
Architecture:Latent diffusion

Essential categories

CategoryStatusNotes
Source code✘ ClosedNo source code.
Training data✘ ClosedDataset is closed and only briefly described.
Model weights✘ ClosedNot provided.
Code documentation✘ ClosedNo source code.
Training procedure✔︎ OpenTraining procedure is properly described.
Evaluation procedure~ PartialMissing data and metrics implementations.
Research paper✔︎ OpenAccepted at ISMIR 2024.
Licensing✘ ClosedNot available.

Desirable categories

CategoryStatusNotes
Model card∅ Not includedNot available.
Datasheet∅ Not includedNot available.
Package∅ Not includedNot available.
User-oriented application∅ Not includedNot available.
Supplementary material page⭐ IncludedA demo page with sound examples and demonstration of the models capabilities is available.

Raw YAML file with complete evaluation.