{"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/divide-and-conquer-a-deep-casa-approach-to","title":"Divide and Conquer: A Deep CASA Approach to Talker-independent Monaural Speaker Separation","arxiv_id":"1904.11148","date":"2019-04-25","proceeding":null,"authors":["Yuzhou Liu","DeLiang Wang"],"abstract":"We address talker-independent monaural speaker separation from the\nperspectives of deep learning and computational auditory scene analysis (CASA).\nSpecifically, we decompose the multi-speaker separation task into the stages of\nsimultaneous grouping and sequential grouping. Simultaneous grouping is first\nperformed in each time frame by separating the spectra of different speakers\nwith a permutation-invariantly trained neural network. In the second stage, the\nframe-level separated spectra are sequentially grouped to different speakers by\na clustering network. The proposed deep CASA approach optimizes frame-level\nseparation and speaker tracking in turn, and produces excellent results for\nboth objectives. Experimental results on the benchmark WSJ0-2mix database show\nthat the new approach achieves the state-of-the-art results with a modest model\nsize.","url_abs":"http://arxiv.org/abs/1904.11148v1","url_pdf":"http://arxiv.org/pdf/1904.11148v1.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":"divide-and-conquer-a-deep-casa-approach-to","repo_url":"https://github.com/yuzhou-git/deep-casa","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"speaker-separation","task_name":"Speaker Separation"},{"task_slug":"speech-separation","task_name":"Speech Separation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/speech-separation-on-wsj0-2mix","task":"Speech Separation","dataset":"WSJ0-2mix","model":"DeepCASA","rank_in_archive_order":30,"of":40,"metrics":{"SI-SDRi":"17.7"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1904.11148","atlas_url":"https://app.syntology.ai/?focus=1904.11148","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}