{"id":33052232,"url":"https://github.com/keunwoochoi/Auralisation","last_synced_at":"2025-11-16T16:02:04.361Z","repository":{"id":114815183,"uuid":"47410461","full_name":"keunwoochoi/Auralisation","owner":"keunwoochoi","description":"Auralisation of learned features in CNN (for audio)","archived":false,"fork":false,"pushed_at":"2017-03-11T14:53:29.000Z","size":72092,"stargazers_count":39,"open_issues_count":0,"forks_count":10,"subscribers_count":5,"default_branch":"master","last_synced_at":"2024-04-22T12:32:37.672Z","etag":null,"topics":[],"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/keunwoochoi.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}},"created_at":"2015-12-04T15:00:53.000Z","updated_at":"2023-11-17T21:07:12.000Z","dependencies_parsed_at":null,"dependency_job_id":"b3f1cc0e-7932-454b-85c2-14c94bf1593d","html_url":"https://github.com/keunwoochoi/Auralisation","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/keunwoochoi/Auralisation","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/keunwoochoi%2FAuralisation","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/keunwoochoi%2FAuralisation/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/keunwoochoi%2FAuralisation/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/keunwoochoi%2FAuralisation/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/keunwoochoi","download_url":"https://codeload.github.com/keunwoochoi/Auralisation/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/keunwoochoi%2FAuralisation/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":284734138,"owners_count":27054622,"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","status":"online","status_checked_at":"2025-11-16T02:00:05.974Z","response_time":65,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":[],"created_at":"2025-11-14T03:00:31.652Z","updated_at":"2025-11-16T16:02:04.344Z","avatar_url":"https://github.com/keunwoochoi.png","language":"Python","funding_links":[],"categories":["Model Explanation"],"sub_categories":["Others","NLP"],"readme":"# AuralisationCNN\nThis repo is for an example of auralisastion of CNNs that is demonstrated on ISMIR 2015.\n\n## Files\nauralise.py: includes all required function for deconvolution.\nexample.py: includes the whole code - just clone and run it by `python example.py`\nYou might need to use older version of Keras, e.g. [this](https://github.com/fchollet/keras/commit/06a1545645d974350d13425246eec53a08cb6ab8) (ver 0.3.x)\n\n## Folders\nsrc_songs: includes three songs that I used in my blog posting.\n\n## Usage\nLoad weights that you want to auralise. I'm using this function\n`W = load_weights()`\nto load my keras model, it can be anything else.\n`W` is a list of weights for the convnet. (TODO: more details)\n\nThen load source files, get STFT of it. I'm using `librosa`.\n\nThen deconve it with `get_deconve_mask`.\n\n## Citation\n[This paper](https://scholar.google.co.kr/citations?view_op=view_citation\u0026hl=en\u0026user=ZrqdSu4AAAAJ\u0026sortby=pubdate\u0026citation_for_view=ZrqdSu4AAAAJ:8k81kl-MbHgC), or simply,\n\n```\n@inproceedings{choi2015auralisation,\n  title={Auralisation of Deep Convolutional Neural Networks: Listening to Learned Features},\n  author={Choi, Keunwoo and Kim, Jeonghee and Fazekas, George and Sandler, Mark},\n  booktitle={International Society of Music Information Retrieval (ISMIR), Late-Breaking/Demo Session, New York, USA},\n  year={2015},\n  organization={International Society of Music Information Retrieval}\n}\n```\n\n## External links\n* [The second blog post](https://keunwoochoi.wordpress.com/2016/03/23/what-cnns-see-when-cnns-see-spectrograms/) has more extensive demo. Detailed description will follow after paper submission.\n* [The first blog post](http://keunwoochoi.blogspot.co.uk/2015/10/ismir-2015-lbd.html) that explains [my ISMIR 2015 Late-Breaking session paper](http://ismir2015.uma.es/LBD/LBD24.pdf).\n\n#### Credits\n* [Keras](https://github.com/fchollet/keras), [librosa](https://bmcfee.github.io/librosa/index.html), [Matt's deconvolution] paper(http://arxiv.org/abs/1311.2901), [Naver Labs](http://labs.naver.com)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkeunwoochoi%2FAuralisation","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fkeunwoochoi%2FAuralisation","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fkeunwoochoi%2FAuralisation/lists"}