OMAR-RQ: Open Music Audio Representation Model Trained with Multi-Feature Masked Token Prediction
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911038949031936 |
|---|---|
| author | Alonso-Jiménez, Pablo Ramoneda, Pedro Araz, R. Oguz Poltronieri, Andrea Bogdanov, Dmitry |
| author_facet | Alonso-Jiménez, Pablo Ramoneda, Pedro Araz, R. Oguz Poltronieri, Andrea Bogdanov, Dmitry |
| contents | Developing open-source foundation models is essential for advancing research in music audio understanding and ensuring access to powerful, multipurpose representations for music information retrieval. We present OMAR-RQ, a model trained with self-supervision via masked token classification methodologies using a large-scale dataset with over 330,000 hours of music audio. We experiment with different input features and quantization options, and achieve state-of-the-art performance in music tagging, pitch estimation, chord recognition, beat tracking, segmentation, and difficulty estimation among open self-supervised models. We open-source our training and evaluation pipelines and model weights, available at https://github.com/mtg/omar-rq. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_03482 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | OMAR-RQ: Open Music Audio Representation Model Trained with Multi-Feature Masked Token Prediction Alonso-Jiménez, Pablo Ramoneda, Pedro Araz, R. Oguz Poltronieri, Andrea Bogdanov, Dmitry Sound Audio and Speech Processing Developing open-source foundation models is essential for advancing research in music audio understanding and ensuring access to powerful, multipurpose representations for music information retrieval. We present OMAR-RQ, a model trained with self-supervision via masked token classification methodologies using a large-scale dataset with over 330,000 hours of music audio. We experiment with different input features and quantization options, and achieve state-of-the-art performance in music tagging, pitch estimation, chord recognition, beat tracking, segmentation, and difficulty estimation among open self-supervised models. We open-source our training and evaluation pipelines and model weights, available at https://github.com/mtg/omar-rq. |
| title | OMAR-RQ: Open Music Audio Representation Model Trained with Multi-Feature Masked Token Prediction |
| topic | Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2507.03482 |