{"id":25215587,"url":"https://github.com/deezer/multi-view-ssl-benchmark","last_synced_at":"2025-04-05T08:43:28.029Z","repository":{"id":233483350,"uuid":"741970695","full_name":"deezer/multi-view-ssl-benchmark","owner":"deezer","description":"Repository for the ICASSP 2024 paper \"An Experimental Comparison Of Multi-view Self-supervised Methods For Music Tagging\".","archived":false,"fork":false,"pushed_at":"2024-05-20T07:43:40.000Z","size":1530,"stargazers_count":6,"open_issues_count":0,"forks_count":1,"subscribers_count":5,"default_branch":"main","last_synced_at":"2025-02-10T18:14:54.582Z","etag":null,"topics":["audio-representation-learning","icassp2024","music-information-retrieval","music-tagging","self-supervised-learning"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/deezer.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-01-11T13:47:24.000Z","updated_at":"2024-12-04T11:25:45.000Z","dependencies_parsed_at":"2024-04-16T11:27:19.907Z","dependency_job_id":"1c405d2c-94b5-44e6-a066-d79a5a25f0a4","html_url":"https://github.com/deezer/multi-view-ssl-benchmark","commit_stats":null,"previous_names":["deezer/multi-view-ssl-benchmark"],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deezer%2Fmulti-view-ssl-benchmark","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deezer%2Fmulti-view-ssl-benchmark/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deezer%2Fmulti-view-ssl-benchmark/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deezer%2Fmulti-view-ssl-benchmark/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/deezer","download_url":"https://codeload.github.com/deezer/multi-view-ssl-benchmark/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247312075,"owners_count":20918341,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["audio-representation-learning","icassp2024","music-information-retrieval","music-tagging","self-supervised-learning"],"created_at":"2025-02-10T18:14:58.778Z","updated_at":"2025-04-05T08:43:27.997Z","avatar_url":"https://github.com/deezer.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# An Experimental Comparison Of Multi-view Self-supervised Methods For Music Tagging\n\nby [Gabriel Meseguer-Brocal](https://www.linkedin.com/in/gabriel-meseguer-brocal-1032a42b), [Dorian Desblancs](https://www.linkedin.com/in/dorian-desblancs), and [Romain Hennequin](http://romain-hennequin.fr/En/index.html).\n\n## About\n\nThis repository contains the models and losses used to generate our results. All trained, self-supervised model weights can be found in the [Releases](https://github.com/deezer/multi-view-ssl-benchmark/releases) section of this repository.\n\n## Getting Started\n\nIn order to explore our repository, one can start with the following:\n```bash\n# Clone and enter repository\ngit clone https://github.com/deezer/multi-view-ssl-benchmark\ncd multi-view-ssl-benchmark\n\n# Install dependencies\npip install poetry\npoetry install\n\n# Download sample audio example\nwget https://github.com/deezer/multi-view-ssl-benchmark/releases/download/v0.0.1/weights.zip\nunzip weights.zip\n```\n\nSample code for model weight loading, audio loading, and embedding computation can then be found in `usage/sample.py`! Use the following command to run it:\n```bash\npoetry run python -m usage.sample\n```\n\n## Other\n\nWe also include the processing of two downstream tasks in `downstream_tasks/`, the `jamendo` and `mtat`. We unfortunately cannot do the same for the Million Song Dataset since it is mapped to songs in the Deezer catalogue.\n\n## Reference\n\nIf you use this repository, please consider citing:\n\n```\n@inproceedings{meseguer2024experimental,\n  title={An Experimental Comparison of Multi-View Self-Supervised Methods for Music Tagging},\n  author={Meseguer-Brocal, Gabriel and Desblancs, Dorian and Hennequin, Romain},\n  booktitle={ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},\n  pages={1141--1145},\n  year={2024},\n  organization={IEEE}\n}\n```\n\nOur paper can be found on [arXiv](https://arxiv.org/abs/2404.09177) 🌟 The poster we presented at ICASSP 2024 can be found in this repo 🗄️\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeezer%2Fmulti-view-ssl-benchmark","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdeezer%2Fmulti-view-ssl-benchmark","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeezer%2Fmulti-view-ssl-benchmark/lists"}