{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T19:57:26Z","timestamp":1776887846997,"version":"3.51.2"},"reference-count":33,"publisher":"IEEE","license":[{"start":{"date-parts":[[2023,6,4]],"date-time":"2023-06-04T00:00:00Z","timestamp":1685836800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2023,6,4]],"date-time":"2023-06-04T00:00:00Z","timestamp":1685836800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,6,4]]},"DOI":"10.1109\/icassp49357.2023.10094836","type":"proceedings-article","created":{"date-parts":[[2023,5,5]],"date-time":"2023-05-05T17:28:30Z","timestamp":1683307710000},"page":"1-5","source":"Crossref","is-referenced-by-count":17,"title":["The Multimodal Information Based Speech Processing (Misp) 2022 Challenge: Audio-Visual Diarization And Recognition"],"prefix":"10.1109","author":[{"given":"Zhe","family":"Wang","sequence":"first","affiliation":[{"name":"University of Science and Technology of China,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shilong","family":"Wu","sequence":"additional","affiliation":[{"name":"University of Science and Technology of China,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hang","family":"Chen","sequence":"additional","affiliation":[{"name":"University of Science and Technology of China,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mao-Kui","family":"He","sequence":"additional","affiliation":[{"name":"University of Science and Technology of China,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jun","family":"Du","sequence":"additional","affiliation":[{"name":"University of Science and Technology of China,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chin-Hui","family":"Lee","sequence":"additional","affiliation":[{"name":"Georgia Institute of Technology,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jingdong","family":"Chen","sequence":"additional","affiliation":[{"name":"Northwestern Polytechnical University,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shinji","family":"Watanabe","sequence":"additional","affiliation":[{"name":"Carnegie Mellon University,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sabato","family":"Siniscalchi","sequence":"additional","affiliation":[{"name":"Georgia Institute of Technology,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Odette","family":"Scharenborg","sequence":"additional","affiliation":[{"name":"Delft University of Technology,The Netherlands"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Diyuan","family":"Liu","sequence":"additional","affiliation":[{"name":"iFlytek,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Baocai","family":"Yin","sequence":"additional","affiliation":[{"name":"iFlytek,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jia","family":"Pan","sequence":"additional","affiliation":[{"name":"iFlytek,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jianqing","family":"Gao","sequence":"additional","affiliation":[{"name":"iFlytek,China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Cong","family":"Liu","sequence":"additional","affiliation":[{"name":"iFlytek,China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2017.2648793"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2011.47"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054376"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.3390\/s19235163"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00369"},{"key":"ref30","doi-asserted-by":"crossref","first-page":"2011","DOI":"10.1109\/TASL.2007.902460","article-title":"Acoustic beamforming for speaker diarization of meetings","volume":"15","author":"anguera","year":"2007","journal-title":"IEEE Transactions on Audio Speech and Language Processing"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1016\/S0167-6393(98)00048-X"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1016\/j.csl.2021.101254"},{"key":"ref10","article-title":"Third dihard challenge evaluation plan","author":"ryant","year":"2020"},{"key":"ref32","article-title":"The kaldi speech recognition toolkit","author":"povey","year":"2011","journal-title":"Proc ASRU"},{"key":"ref2","article-title":"Deep audio-visual speech recognition","author":"afouras","year":"2018","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1467-8721.2008.00615.x"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2018.2815268"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414567"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1007\/11965152_28"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10483"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10106"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.21437\/CHiME.2020-1"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-3015"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053841"},{"key":"ref20","doi-asserted-by":"crossref","first-page":"2616","DOI":"10.21437\/Interspeech.2017-950","article-title":"VoxCeleb: A Large-Scale Speaker Identification Dataset","author":"nagrani","year":"2017","journal-title":"Proc INTERSPEECH"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054017"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1929"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747067"},{"key":"ref27","first-page":"1","article-title":"Nara-wpe: A python package for weighted prediction error dereverberation in numpy and tensorflow for online and offline processing","author":"drude","year":"2018","journal-title":"Speech Communication 13th ITG-Symposium"},{"key":"ref29","first-page":"881","article-title":"Dover-lap: A method for combining overlap-aware diariza- tion outputs","author":"raj","year":"2021","journal-title":"Proc SLT"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8461375"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.21437\/CHiME.2020-11"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.21437\/CHiME.2020-9"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746683"},{"key":"ref3","first-page":"6447","article-title":"Lip reading sentences in the wild","author":"chung","year":"2017","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746874"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747535"}],"event":{"name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Rhodes Island, Greece","start":{"date-parts":[[2023,6,4]]},"end":{"date-parts":[[2023,6,10]]}},"container-title":["ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10094559\/10094560\/10094836.pdf?arnumber=10094836","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,20]],"date-time":"2023-11-20T18:59:06Z","timestamp":1700506746000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10094836\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,6,4]]},"references-count":33,"URL":"https:\/\/doi.org\/10.1109\/icassp49357.2023.10094836","relation":{},"subject":[],"published":{"date-parts":[[2023,6,4]]}}}