{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T07:12:03Z","timestamp":1775200323293,"version":"3.50.1"},"reference-count":36,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,12,6]],"date-time":"2025-12-06T00:00:00Z","timestamp":1764979200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,12,6]],"date-time":"2025-12-06T00:00:00Z","timestamp":1764979200000},"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":[[2025,12,6]]},"DOI":"10.1109\/asru65441.2025.11434798","type":"proceedings-article","created":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T19:48:04Z","timestamp":1775159284000},"page":"1-7","source":"Crossref","is-referenced-by-count":0,"title":["Predictive ASR and Turn-taking Prediction at Once: Towards More Responsive Spoken Dialog System"],"prefix":"10.1109","author":[{"given":"Ryo","family":"Fukuda","sequence":"first","affiliation":[{"name":"NTT, Inc,Japan"}]},{"given":"Takatomo","family":"Kano","sequence":"additional","affiliation":[{"name":"NTT, Inc,Japan"}]},{"given":"Naohiro","family":"Tawara","sequence":"additional","affiliation":[{"name":"NTT, Inc,Japan"}]},{"given":"Marc","family":"Delcroix","sequence":"additional","affiliation":[{"name":"NTT, Inc,Japan"}]},{"given":"Atsunori","family":"Ogawa","sequence":"additional","affiliation":[{"name":"NTT, Inc,Japan"}]},{"given":"Yuya","family":"Chiba","sequence":"additional","affiliation":[{"name":"NTT, Inc,Japan"}]},{"given":"Atsushi","family":"Ando","sequence":"additional","affiliation":[{"name":"NTT, Inc,Japan"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.2307\/412243"},{"key":"ref2","first-page":"125","article-title":"Utterance units in spoken dialogue","volume-title":"Workshop on dialogue processing in spoken language systems","author":"Traum"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyg.2015.00731"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1016\/j.cognition.2002.06.001"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-211"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-298"},{"key":"ref7","article-title":"Predictive speech recognition and end-of-utterance detection towards spoken dialog systems","author":"Zink","year":"2024","journal-title":"arXiv preprint arXiv:2409.19990"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10955"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W17-5527"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/3242969.3242997"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyg.2012.00376"},{"key":"ref12","article-title":"Robust speech recognition via large-scale weak supervision","author":"Radford","year":"2022","journal-title":"arXiv preprint arXiv:2212.04356"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49660.2025.10889314"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1992.225858"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-67"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.11159\/jmids.2024.017"},{"key":"ref17","article-title":"Sequence transduction with recurrent neural networks","volume-title":"Proceedings of Representation Learning Workshop on Interna- tional Conference on Machine Learning (ICML)","author":"Graves"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-3015"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1016\/j.csl.2020.101178"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.21437\/ICSLP.2002-293"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2006-550"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-1593"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3340555.3353727"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.findings-emnlp.268"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-5024"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-566"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-971"},{"key":"ref29","article-title":"Talking turns: Benchmarking audio foundation models on turn-taking dynamics","volume-title":"The Thirteenth International Conference on Learning Representations (ICLR)","author":"Arora"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.21437\/ICSLP.1998-588"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10447196"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-1456"},{"key":"ref33","article-title":"Decoupled weight decay regularization","volume-title":"The Seventh International Conference on Learning Representations (ICLR)","author":"Loshchilov"},{"issue":"3","key":"ref34","first-page":"13","article-title":"The roles of syntactic and prosodic features in turn-taking: A quantitative analysis of three-party conversations in japanese","volume":"14","author":"Koiso","year":"2010","journal-title":"Journal of the Phonetic Society of Japan"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-3152"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2018-2124"}],"event":{"name":"2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","location":"Honolulu, HI, USA","start":{"date-parts":[[2025,12,6]]},"end":{"date-parts":[[2025,12,10]]}},"container-title":["2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11434577\/11433836\/11434798.pdf?arnumber=11434798","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T04:59:55Z","timestamp":1775192395000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11434798\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,6]]},"references-count":36,"URL":"https:\/\/doi.org\/10.1109\/asru65441.2025.11434798","relation":{},"subject":[],"published":{"date-parts":[[2025,12,6]]}}}