MUSIC GENRE ANALYSER - THIRD-PARTY NOTICES

Discogs-EffNet, Music Technology Group, Universitat Pompeu Fabra.
Model: discogs-effnet-bsdynamic-1.onnx, 400 Discogs styles.
Public model licence: CC BY-NC-SA 4.0.
The site operator has obtained separate commercial authorization for its use
of this model. That authorization does not grant visitors redistribution or
commercial rights to the model files. The public licence remains available at:
https://essentia.upf.edu/models.html
https://creativecommons.org/licenses/by-nc-sa/4.0/
Alonso-Jimenez, Serra and Bogdanov (2022), Music Representation Learning
Based on Editorial Metadata from Discogs, ISMIR.

Essentia.js 0.1.3 - AGPL-3.0.
Licence: essentia-LICENSE.txt (same directory).
Source: https://github.com/MTG/essentia.js
The distributed feature-extraction library is unmodified.
This page's corresponding integration source is available in this project's
src/lib/musicAnalysis, src/components/music-analysis and scripts/prepare-music-analysis.mjs.

ONNX Runtime Web - MIT.
Source and licence: https://github.com/microsoft/onnxruntime

Audio is decoded and analysed in the browser. The model and runtime files are
downloaded from this site's static assets. No audio upload endpoint is used.
Scores are independent model outputs, not calibrated confidence estimates.