{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/training-neural-audio-classifiers-with-few","title":"Training neural audio classifiers with few data","arxiv_id":"1810.10274","date":"2018-10-24","proceeding":null,"authors":["Jordi Pons","Joan Serrà","Xavier Serra"],"abstract":"We investigate supervised learning strategies that improve the training of\nneural network audio classifiers on small annotated collections. In particular,\nwe study whether (i) a naive regularization of the solution space, (ii)\nprototypical networks, (iii) transfer learning, or (iv) their combination, can\nfoster deep learning models to better leverage a small amount of training\nexamples. To this end, we evaluate (i-iv) for the tasks of acoustic event\nrecognition and acoustic scene classification, considering from 1 to 100\nlabeled examples per class. Results indicate that transfer learning is a\npowerful strategy in such scenarios, but prototypical networks show promising\nresults when one does not count with external or validation data.","url_abs":"http://arxiv.org/abs/1810.10274v3","url_pdf":"http://arxiv.org/pdf/1810.10274v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"training-neural-audio-classifiers-with-few","repo_url":"https://github.com/jordipons/neural-classifiers-with-few-audio","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"training-neural-audio-classifiers-with-few","repo_url":"https://github.com/yagyapandeya/CNN-with-Few-Data-VGGish-","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"acoustic-scene-classification","task_name":"Acoustic Scene Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"scene-classification","task_name":"Scene Classification"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}