{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,19]],"date-time":"2026-07-19T09:49:13Z","timestamp":1784454553951,"version":"3.55.0"},"reference-count":45,"publisher":"IEEE","license":[{"start":{"date-parts":[[2024,12,2]],"date-time":"2024-12-02T00:00:00Z","timestamp":1733097600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,12,2]],"date-time":"2024-12-02T00:00:00Z","timestamp":1733097600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2024,12,2]]},"DOI":"10.1109\/slt61566.2024.10832163","type":"proceedings-article","created":{"date-parts":[[2025,1,16]],"date-time":"2025-01-16T18:31:27Z","timestamp":1737052287000},"page":"294-301","source":"Crossref","is-referenced-by-count":6,"title":["DDTSE: Discriminative Diffusion Model for Target Speech Extraction"],"prefix":"10.1109","author":[{"given":"Leying","family":"Zhang","sequence":"first","affiliation":[{"name":"Shanghai Jiao Tong University,Auditory Cognition and Computational Acoustics Lab MoE Key Lab of Artificial Intelligence, AI Institute,Department of Computer Science and Engineering,Shanghai,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yao","family":"Qian","sequence":"additional","affiliation":[{"name":"Microsoft,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Linfeng","family":"Yu","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University,Auditory Cognition and Computational Acoustics Lab MoE Key Lab of Artificial Intelligence, AI Institute,Department of Computer Science and Engineering,Shanghai,China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Heming","family":"Wang","sequence":"additional","affiliation":[{"name":"Microsoft,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hemin","family":"Yang","sequence":"additional","affiliation":[{"name":"Microsoft,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shujie","family":"Liu","sequence":"additional","affiliation":[{"name":"Microsoft,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Long","family":"Zhou","sequence":"additional","affiliation":[{"name":"Microsoft,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanmin","family":"Qian","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University,Auditory Cognition and Computational Acoustics Lab MoE Key Lab of Artificial Intelligence, AI Institute,Department of Computer Science and Engineering,Shanghai,China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"issue":"1","key":"ref1","first-page":"117","article-title":"The cocktail party phenomenon: A review of research on speech intelligibility in multiple-talker conditions","volume":"86","author":"Bronkhorst","year":"2000","journal-title":"Acta Acustica united with Acustica"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2023.3240008"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1631\/FITEE.1700814"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462661"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2022-645"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1101"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10095258"},{"key":"ref8","first-page":"7292","article-title":"Dpcen: Densely-connected pyramid complex convolutional network for robust speech separation and extraction","volume-title":"ICASSP","author":"Han"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2020.2987429"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2019.2922820"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2023.3294692"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2023.3285241"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10446418"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2020.3025020"},{"key":"ref15","article-title":"Dose: Diffusion dropout with adaptive prior for speech enhancement","volume":"36","author":"Tai","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10653"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-1234"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-1445"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096064"},{"key":"ref20","article-title":"Conditional diffusion model for target speaker extraction","author":"Nguyen","year":"2023","journal-title":"arXiv preprint arXiv:2310.04791"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1017\/9781108186735"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1016\/0304-4149(82)90051-5"},{"key":"ref23","article-title":"Score-based generative modeling through stochastic differential equations","volume-title":"International Conference on Learning Representations","author":"Song"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1162\/NECO_a_00142"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2023-1547"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10448429"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10446805"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10447860"},{"key":"ref29","first-page":"6840","article-title":"Denoising diffusion probabilistic models","volume":"33","author":"Ho","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref30","article-title":"Understanding the diffusion models by conditional expectations","author":"Lu","year":"2023","journal-title":"arXiv preprint arXiv:2301.07882"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2017.12.012"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11671"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10095310"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096626"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-2537"},{"key":"ref36","article-title":"Librimix: An opensource dataset for generalizable speech separation","author":"Cosentino","year":"2020","journal-title":"arXiv preprint arXiv:2005.11262"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2821"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/WASPAA.2019.8937202"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2001.941023"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2016.2585878"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8683855"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414878"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746108"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1145\/3503161.3547855"}],"event":{"name":"2024 IEEE Spoken Language Technology Workshop (SLT)","location":"Macao","start":{"date-parts":[[2024,12,2]]},"end":{"date-parts":[[2024,12,5]]}},"container-title":["2024 IEEE Spoken Language Technology Workshop (SLT)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10830790\/10830793\/10832163.pdf?arnumber=10832163","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,1,17]],"date-time":"2025-01-17T08:19:06Z","timestamp":1737101946000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10832163\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,12,2]]},"references-count":45,"URL":"https:\/\/doi.org\/10.1109\/slt61566.2024.10832163","relation":{},"subject":[],"published":{"date-parts":[[2024,12,2]]}}}