{"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/ts-sep-joint-diarization-and-separation","title":"TS-SEP: Joint Diarization and Separation Conditioned on Estimated Speaker Embeddings","arxiv_id":"2303.03849","date":"2023-03-07","proceeding":null,"authors":["Christoph Boeddeker","Aswin Shanmugam Subramanian","Gordon Wichern","Reinhold Haeb-Umbach","Jonathan Le Roux"],"abstract":"Since diarization and source separation of meeting data are closely related tasks, we here propose an approach to perform the two objectives jointly. It builds upon the target-speaker voice activity detection (TS-VAD) diarization approach, which assumes that initial speaker embeddings are available. We replace the final combined speaker activity estimation network of TS-VAD with a network that produces speaker activity estimates at a time-frequency resolution. Those act as masks for source extraction, either via masking or via beamforming. The technique can be applied both for single-channel and multi-channel input and, in both cases, achieves a new state-of-the-art word error rate (WER) on the LibriCSS meeting data recognition task. We further compute speaker-aware and speaker-agnostic WERs to isolate the contribution of diarization errors to the overall WER performance.","url_abs":"https://arxiv.org/abs/2303.03849v3","url_pdf":"https://arxiv.org/pdf/2303.03849v3.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":"ts-sep-joint-diarization-and-separation","repo_url":"https://github.com/merlresearch/tssep","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"action-detection","task_name":"Action Detection"},{"task_slug":"activity-detection","task_name":"Activity Detection"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/speech-recognition-on-libricss","task":"Speech Recognition","dataset":"LibriCSS","model":"TS-SEP","rank_in_archive_order":1,"of":2,"metrics":{"Word Error Rate (WER)":"3.27"},"uses_additional_data":true}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}