Add comprehensive model card for OMAR-RQ

#1
by nielsr HF Staff - opened

This PR adds a comprehensive model card for the OMAR-RQ model, based on the paper OMAR-RQ: Open Music Audio Representation Model Trained with Multi-Feature Masked Token Prediction.

It includes the following improvements:

  • Adds pipeline_tag: audio-classification for discoverability (e.g., at https://huggingface.co/models?pipeline_tag=audio-classification).
  • Specifies library_name: omar_rq for proper library attribution.
  • Explicitly states the license: cc-by-nc-sa-4.0 for model weights.
  • Provides a link to the original paper and the GitHub repository (https://github.com/MTG/OMAR-RQ).
  • Includes a clear sample usage section for inference.
  • Adds information about available models and their performance, along with Hugging Face Model IDs.
  • Includes detailed licensing information and a citation section.

Please review and merge this PR if everything looks good.

Music Technology Group (Universitat Pompeu Fabra) org

@nielsr ,
Thank you very much for your contribution.
I committed a version of your model card suggestion with the changes applied in
https://huggingface.co/mtg-upf/omar-rq-multifeature-25hz-fsq/discussions/1#686d51a0641747b25a1bffb2

p-alonso changed pull request status to closed

Sign up or log in to comment