{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T16:19:38Z","timestamp":1781713178641,"version":"3.54.5"},"reference-count":23,"publisher":"IEEE","license":[{"start":{"date-parts":[[2022,8,7]],"date-time":"2022-08-07T00:00:00Z","timestamp":1659830400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,8,7]],"date-time":"2022-08-07T00:00:00Z","timestamp":1659830400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100010623","name":"University of Thessaly","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100010623","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2022,8,7]]},"DOI":"10.1109\/mwscas54063.2022.9859533","type":"proceedings-article","created":{"date-parts":[[2022,8,22]],"date-time":"2022-08-22T20:02:33Z","timestamp":1661198553000},"page":"1-4","source":"Crossref","is-referenced-by-count":3,"title":["Audio-visual Speaker Diarization: Improved Voice Activity Detection with CNN based Feature Extraction"],"prefix":"10.1109","author":[{"given":"Konstantinos","family":"Fanaras","sequence":"first","affiliation":[{"name":"King Abdullah University of Science and Technology (KAUST),Innovative Technologies Laboratories (ITL),Thuwal,Saudi Arabia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Antonios","family":"Tragoudaras","sequence":"additional","affiliation":[{"name":"King Abdullah University of Science and Technology (KAUST),Innovative Technologies Laboratories (ITL),Thuwal,Saudi Arabia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Charalampos","family":"Antoniadis","sequence":"additional","affiliation":[{"name":"King Abdullah University of Science and Technology (KAUST),Innovative Technologies Laboratories (ITL),Thuwal,Saudi Arabia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yehia","family":"Massoud","sequence":"additional","affiliation":[{"name":"King Abdullah University of Science and Technology (KAUST),Innovative Technologies Laboratories (ITL),Thuwal,Saudi Arabia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref10","article-title":"Mfcc and its applications in speaker recognition","author":"tiwari","year":"2010","journal-title":"International Journal on Emerging Technologies"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/SLT.2018.8639585"},{"key":"ref12","article-title":"End-to-end domain-adversarial voice activity detection","author":"lavechin","year":"2019","journal-title":"arXiv preprint arXiv 1910 10335"},{"key":"ref13","first-page":"28","article-title":"The AMI meeting corpus: A pre-announcement","author":"j c","year":"2005","journal-title":"Proc Int Conf Methods and Techn Behavioral Res"},{"key":"ref14","first-page":"192","article-title":"S3 FD: Single shot scale-invariant face detector","author":"zhang","year":"2017","journal-title":"ICCV"},{"key":"ref15","article-title":"Parsenet: Looking wider to see better","author":"liu","year":"2015","journal-title":"arXiv 1506 04579"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.596"},{"key":"ref17","article-title":"Very deep convolutional networks for large-scale image recognition","author":"simonyan","year":"2015","journal-title":"ICLRE"},{"key":"ref18","first-page":"249","article-title":"Understanding the difficulty of training deep feedforward neural networks","author":"glorot","year":"2010","journal-title":"Proc AISTATS"},{"key":"ref19","first-page":"593","article-title":"Good features to track","author":"shi","year":"1994","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1145\/3474085.3475587"},{"key":"ref3","article-title":"Out of time: automated lip sync in the wild","author":"chung","year":"2016","journal-title":"Workshop on Multi-view Lip-reading ACCV"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2010-704"},{"key":"ref5","first-page":"4069","article-title":"Multimodal speaker diarization of real-world meetings using compressed-domain video features","author":"friedland","year":"2009","journal-title":"Proc ICASSP"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.4304\/jmm.5.4.322-331"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2009-338"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-3116"},{"key":"ref9","first-page":"33","article-title":"Speaker Identification and Verification using Vector Quantization and Mel Frequency Cepstral Coefficients","volume":"4","author":"srinivasan","year":"2012","journal-title":"Research Journal of Applied Sciences Engineering and Technology"},{"key":"ref1","article-title":"AVAAVD: Audio-visual Speaker Diarization in the Wild","author":"xu","year":"2021","journal-title":"arXiv preprint arXiv 2111 14382"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1007\/BF01890115"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.25080\/Majora-7b98e3ed-003"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.5244\/C.28.6"}],"event":{"name":"2022 IEEE 65th International Midwest Symposium on Circuits and Systems (MWSCAS)","location":"Fukuoka, Japan","start":{"date-parts":[[2022,8,7]]},"end":{"date-parts":[[2022,8,10]]}},"container-title":["2022 IEEE 65th International Midwest Symposium on Circuits and Systems (MWSCAS)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9859262\/9859267\/09859533.pdf?arnumber=9859533","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,9,12]],"date-time":"2022-09-12T19:57:49Z","timestamp":1663012669000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9859533\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,7]]},"references-count":23,"URL":"https:\/\/doi.org\/10.1109\/mwscas54063.2022.9859533","relation":{},"subject":[],"published":{"date-parts":[[2022,8,7]]}}}