{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T12:08:41Z","timestamp":1773144521673,"version":"3.50.1"},"reference-count":35,"publisher":"IEEE","license":[{"start":{"date-parts":[[2022,10,18]],"date-time":"2022-10-18T00:00:00Z","timestamp":1666051200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2022,10,18]],"date-time":"2022-10-18T00:00:00Z","timestamp":1666051200000},"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":[[2022,10,18]]},"DOI":"10.1109\/acii55700.2022.9953889","type":"proceedings-article","created":{"date-parts":[[2022,11,25]],"date-time":"2022-11-25T17:41:10Z","timestamp":1669398070000},"page":"1-7","source":"Crossref","is-referenced-by-count":3,"title":["Audio and ASR-based Filled Pause Detection"],"prefix":"10.1109","author":[{"given":"Aggelina","family":"Chatziagapi","sequence":"first","affiliation":[{"name":"Stony Brook University & Behavioral Signal Technologies,New York,U.S.A"}]},{"given":"Dimitris","family":"Sgouropoulos","sequence":"additional","affiliation":[{"name":"Behavioral Signal Technologies,Athens,Greece"}]},{"given":"Constantinos","family":"Karouzos","sequence":"additional","affiliation":[{"name":"Behavioral Signal Technologies,Athens,Greece"}]},{"given":"Thomas","family":"Melistas","sequence":"additional","affiliation":[{"name":"Behavioral Signal Technologies,Athens,Greece"}]},{"given":"Theodoros","family":"Giannakopoulos","sequence":"additional","affiliation":[{"name":"Behavioral Signal Technologies & NCSR Demokritos,Athens,Greece"}]},{"given":"Athanasios","family":"Katsamanis","sequence":"additional","affiliation":[{"name":"Behavioral Signal Technologies,Athens,Greece"}]},{"given":"Shrikanth","family":"Narayanan","sequence":"additional","affiliation":[{"name":"Behavioral Signal Technologies,Los Angeles,U.S.A"}]}],"member":"263","reference":[{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1007\/s10579-010-9120-1"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-3243"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-3201"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00949"},{"key":"ref35","article-title":"Attention is all you need","volume":"30","author":"vaswani","year":"2017","journal-title":"Advances in neural information processing systems"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.3115\/1220575.1220605"},{"key":"ref10","author":"inaguma","year":"2017","journal-title":"Social signal detection in spontaneous dialogue using bidirectional 1stm-ctc"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-635"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-1247"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1992.225858"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D17-1296"},{"key":"ref15","first-page":"3754","article-title":"Improving disfluency detection by self-training a self-attentive model","author":"lou","year":"0","journal-title":"Proceedings annual meeting of the Association for Computational Linguistics"},{"key":"ref16","first-page":"4171","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"devlin","year":"2019","journal-title":"Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics Human Language Technologies Volume 1 (Long and Short Papers)"},{"key":"ref17","first-page":"834","article-title":"Video and audio based detection of filled hesitation pauses in classroom lectures","author":"tsiaras","year":"0","journal-title":"17th European Signal Processing Conference 2009"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D16-1058"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2998"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.3390\/app8010150"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.21437\/Eurospeech.1999-60"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2561"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1016\/j.cognition.2006.10.010"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-1528"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2016.2528162"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.21437\/SSW.2019-44"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2013-61"},{"key":"ref7","first-page":"1","article-title":"Accelerating manual annotation of filled pauses by automatic preselection","author":"egorow","year":"0","journal-title":"2017 International Conference on Companion Technology (ICCT) ICCT"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1163\/22105832-00602001"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/N18-1007"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2013-56"},{"key":"ref20","article-title":"Neural machine translation by jointly learning to align and translate","author":"bahdanau","year":"0","journal-title":"3rd International Conference on Learning Representations ICLR 2015"},{"key":"ref22","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","author":"krizhevsky","year":"2012","journal-title":"Advances in Neural Infor-Mation Processing Systems"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1162"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7952585"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2014.2339736"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2017.2766843"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2018.05.016"}],"event":{"name":"2022 10th International Conference on Affective Computing and Intelligent Interaction (ACII)","location":"Nara, Japan","start":{"date-parts":[[2022,10,18]]},"end":{"date-parts":[[2022,10,21]]}},"container-title":["2022 10th International Conference on Affective Computing and Intelligent Interaction (ACII)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/9953736\/9953807\/09953889.pdf?arnumber=9953889","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,12,19]],"date-time":"2022-12-19T15:03:46Z","timestamp":1671462226000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/9953889\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,10,18]]},"references-count":35,"URL":"https:\/\/doi.org\/10.1109\/acii55700.2022.9953889","relation":{},"subject":[],"published":{"date-parts":[[2022,10,18]]}}}