{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T16:19:39Z","timestamp":1781713179712,"version":"3.54.5"},"reference-count":33,"publisher":"IEEE","license":[{"start":{"date-parts":[[2024,4,14]],"date-time":"2024-04-14T00:00:00Z","timestamp":1713052800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,4,14]],"date-time":"2024-04-14T00:00:00Z","timestamp":1713052800000},"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":[[2024,4,14]]},"DOI":"10.1109\/icasspw62465.2024.10626914","type":"proceedings-article","created":{"date-parts":[[2024,8,15]],"date-time":"2024-08-15T17:19:18Z","timestamp":1723742358000},"page":"174-178","source":"Crossref","is-referenced-by-count":3,"title":["Late Audio-Visual Fusion for in-the-Wild Speaker Diarization"],"prefix":"10.1109","author":[{"given":"Zexu","family":"Pan","sequence":"first","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gordon","family":"Wichern","sequence":"additional","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fran\u00e7ois G.","family":"Germain","sequence":"additional","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aswin","family":"Subramanian","sequence":"additional","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jonathan Le","family":"Roux","sequence":"additional","affiliation":[{"name":"Mitsubishi Electric Research Laboratories (MERL),Cambridge,MA,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9413776"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-157"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1653"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1016\/j.csl.2021.101254"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1016\/j.csl.2021.101317"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1022"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2022.3162080"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-1176"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7952154"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2022.3205759"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2022.3175130"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-3116"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3531232.3531250"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10106"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2899"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1145\/3503161.3548027"},{"key":"ref17","article-title":"Third DIHARD challenge evaluation plan","author":"Ryant","year":"2020"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/MMSP55362.2022.9948860"},{"key":"ref19","article-title":"Is someone speaking? Exploring long-term temporal features for audio-visual active speaker detection","volume-title":"Proc. ACM Multimedia","author":"Tao"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ASYU50717.2020.9259802"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298682"},{"key":"ref22","article-title":"Learning speaker representation for neural network based multichannel speaker extraction","volume-title":"Proc. ASRU","author":"Zmol \u02c7 \u0131kov\u00e1"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414023"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2022.3153258"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414508"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1007\/BF01890115"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053900"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1929"},{"key":"ref29","article-title":"MUSAN: A music, speech, and noise corpus","author":"Snyder","year":"2015"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3133208"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746005"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-126"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1007\/11677482_36"}],"event":{"name":"2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)","location":"Seoul, Korea, Republic of","start":{"date-parts":[[2024,4,14]]},"end":{"date-parts":[[2024,4,19]]}},"container-title":["2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10625769\/10625780\/10626914.pdf?arnumber=10626914","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,16]],"date-time":"2024-08-16T05:53:06Z","timestamp":1723787586000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10626914\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,4,14]]},"references-count":33,"URL":"https:\/\/doi.org\/10.1109\/icasspw62465.2024.10626914","relation":{},"subject":[],"published":{"date-parts":[[2024,4,14]]}}}