{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,29]],"date-time":"2025-11-29T06:51:33Z","timestamp":1764399093143,"version":"3.46.0"},"reference-count":35,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T00:00:00Z","timestamp":1761091200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T00:00:00Z","timestamp":1761091200000},"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":[[2025,10,22]]},"DOI":"10.1109\/apsipaasc65261.2025.11249303","type":"proceedings-article","created":{"date-parts":[[2025,11,28]],"date-time":"2025-11-28T18:40:26Z","timestamp":1764355226000},"page":"1057-1061","source":"Crossref","is-referenced-by-count":0,"title":["Speech Intelligibility Assessment with Uncertainty-Aware Whisper Embeddings and sLSTM"],"prefix":"10.1109","author":[{"given":"Ryandhimas E.","family":"Zezario","sequence":"first","affiliation":[{"name":"Academia Sinica,Research Center for Information Technology Innovation,Taipei,Taiwan"}]},{"given":"Dyah A.M.G.","family":"Wisnu","sequence":"additional","affiliation":[{"name":"Institute of Information Science, Academia Sinica,Social Network and Human Centered Computing,Taipei,Taiwan"}]},{"given":"Hsin-Min","family":"Wang","sequence":"additional","affiliation":[{"name":"Institute of Information Science, Academia Sinica,Taipei,Taiwan"}]},{"given":"Yu","family":"Tsao","sequence":"additional","affiliation":[{"name":"Academia Sinica,Research Center for Information Technology Innovation,Taipei,Taiwan"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1201\/9781420015836"},{"key":"ref2","first-page":"862","article-title":"Perceptual evaluation of speech quality (PESQ), an objective method for end-to-end speech quality assessment of narrow-band telephone networks and speech codecs","author":"Rix","year":"2001","journal-title":"in ITU-T Recommendation"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1016\/j.specom.2020.05.001"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.17743\/jaes.2014.0006"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1121\/1.384464"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1121\/1.1916407"},{"journal-title":"in Acoustical Society of America","article-title":"Methods for calculation of the speech intelligibility index","year":"1997","key":"ref7"},{"issue":"6","key":"ref8","first-page":"355","article-title":"Evaluation of speech transmission channels by using artificial signals","volume":"25","author":"Houtgast","year":"1971","journal-title":"Acustica"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/tasl.2011.2114881"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9052949"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2018.2847459"},{"key":"ref12","first-page":"482","article-title":"STOI-Net: A deep learning-based nonintrusive speech intelligibility assessment model","author":"Zezario","year":"2020","journal-title":"in Proc. APSIPA ASC"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1121\/10.0034362"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10399"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10828"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2022.3205757"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10838"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10447907"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/icassp48485.2024.10447597"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICME57554.2024.10688047"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1016\/j.apacoust.2023.109663"},{"key":"ref22","first-page":"112","article-title":"Wav2vec 2.0: A framework for self-supervised learning of speech representations","author":"Baevski","year":"2020","journal-title":"in Proc. NIPS"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3122291"},{"key":"ref24","first-page":"28492","article-title":"Robust speech recognition via large-scale weak supervision","author":"Radford","year":"2023","journal-title":"in Proc. ICML"},{"key":"ref25","first-page":"1","article-title":"XLSTM: Extended long short-term memory","author":"Beck","year":"2024","journal-title":"in Proc. NeurIPS"},{"key":"ref26","article-title":"Speech recognition (version 3.6) [software]","author":"Zhang","year":"2017","journal-title":"in Proc. ICCC"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/SLT.2018.8639585"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10153"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU57964.2023.10389763"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/TASSP.1985.1164550"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/APSIPA.2017.8281993"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053591"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-517"},{"journal-title":"Music source separation in the waveform domain","year":"2021","author":"D\u00e9fossez","key":"ref34"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.2307\/1412159"}],"event":{"name":"2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","start":{"date-parts":[[2025,10,22]]},"location":"Singapore, Singapore","end":{"date-parts":[[2025,10,24]]}},"container-title":["2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11248853\/11248968\/11249303.pdf?arnumber=11249303","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,29]],"date-time":"2025-11-29T06:50:43Z","timestamp":1764399043000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11249303\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,22]]},"references-count":35,"URL":"https:\/\/doi.org\/10.1109\/apsipaasc65261.2025.11249303","relation":{},"subject":[],"published":{"date-parts":[[2025,10,22]]}}}