{"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/convolutional-neural-network-achieves-human","title":"Convolutional Neural Network Achieves Human-level Accuracy in Music Genre Classification","arxiv_id":"1802.09697","date":"2018-02-27","proceeding":null,"authors":["Mingwen Dong"],"abstract":"Music genre classification is one example of content-based analysis of music\nsignals. Traditionally, human-engineered features were used to automatize this\ntask and 61% accuracy has been achieved in the 10-genre classification.\nHowever, it's still below the 70% accuracy that humans could achieve in the\nsame task. Here, we propose a new method that combines knowledge of human\nperception study in music genre classification and the neurophysiology of the\nauditory system. The method works by training a simple convolutional neural\nnetwork (CNN) to classify a short segment of the music signal. Then, the genre\nof a music is determined by splitting it into short segments and then combining\nCNN's predictions from all short segments. After training, this method achieves\nhuman-level (70%) accuracy and the filters learned in the CNN resemble the\nspectrotemporal receptive field (STRF) in the auditory system.","url_abs":"http://arxiv.org/abs/1802.09697v1","url_pdf":"http://arxiv.org/pdf/1802.09697v1.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":"convolutional-neural-network-achieves-human","repo_url":"https://github.com/ds7711/music_genre_classification","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"convolutional-neural-network-achieves-human","repo_url":"https://github.com/belqiliass/Music-Classifier","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"genre-classification","task_name":"Genre classification"},{"task_slug":"music-genre-classification","task_name":"Music Genre Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}