{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,21]],"date-time":"2025-02-21T01:36:03Z","timestamp":1740101763191,"version":"3.37.3"},"reference-count":49,"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:\/\/summer-heart-0930.chufeiyun1688.workers.dev:443\/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:\/\/summer-heart-0930.chufeiyun1688.workers.dev:443\/https\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100006180","name":"Technology Development","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006180","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\/icassp49357.2023.10097071","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":0,"title":["Intermpl: Momentum Pseudo-Labeling With Intermediate CTC Loss"],"prefix":"10.1109","author":[{"given":"Yosuke","family":"Higuchi","sequence":"first","affiliation":[{"name":"Carnegie Mellon University,USA"}]},{"given":"Tetsuji","family":"Ogawa","sequence":"additional","affiliation":[{"name":"Waseda University,Japan"}]},{"given":"Tetsunori","family":"Kobayashi","sequence":"additional","affiliation":[{"name":"Waseda University,Japan"}]},{"given":"Shinji","family":"Watanabe","sequence":"additional","affiliation":[{"name":"Carnegie Mellon University,USA"}]}],"member":"263","reference":[{"key":"ref13","first-page":"449","article-title":"A comparative study on Transformer vs RNN in speech applications","author":"karita","year":"2019","journal-title":"Proc ASRU"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1780"},{"key":"ref15","article-title":"Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks","author":"lee","year":"2013","journal-title":"Proc ICML"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9052942"},{"key":"ref11","first-page":"4774","article-title":"State-ofthe-art speech recognition with sequence-to-sequence models","author":"chiu","year":"2018","journal-title":"Proc ICASSP"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-3015"},{"key":"ref17","first-page":"7054","article-title":"Sequencelevel consistency training for semi-supervised end-to-end automatic speech recognition","author":"masumura","year":"2020","journal-title":"Proc ICASSP"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682172"},{"article-title":"Semisupervised speech recognition via local prior matching","year":"2020","author":"hsu","key":"ref19"},{"key":"ref18","article-title":"Semisupervised learning with data augmentation for end-to-end ASR","author":"weninger","year":"2020","journal-title":"Proc INTERSPEECH"},{"key":"ref46","article-title":"Adam: A method for stochastic optimization","author":"kingma","year":"2015","journal-title":"Proc ICLR"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414858"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2008-122"},{"key":"ref47","article-title":"Attention is all you need","author":"vaswani","year":"2017","journal-title":"Proc NeurIPS"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-99579-3_21"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P18-1007"},{"key":"ref43","article-title":"The Kaldi speech recognition toolkit","author":"povey","year":"2011","journal-title":"Proc ASRU"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-11276"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462506"},{"key":"ref7","article-title":"Neural machine translation by jointly learning to align and translate","author":"bahdanau","year":"2014","journal-title":"Proc ICLR"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053889"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1145\/1143844.1143891"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472621"},{"key":"ref6","first-page":"3104","article-title":"Sequence to sequence learning with neural networks","author":"sutskever","year":"2014","journal-title":"Proc NeurIPS"},{"article-title":"Sequence transduction with recurrent neural networks","year":"2012","author":"graves","key":"ref5"},{"key":"ref40","article-title":"ESPnet: End-toend speech processing toolkit","author":"watanabe","year":"2018","journal-title":"Proc INTERSPEECH"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-911"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU51503.2021.9688157"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/SLT.2018.8639530"},{"key":"ref36","first-page":"7797","article-title":"Hierarchical conditional end-to-end ASR with CTC and multi-granular subword units","author":"higuchi","year":"2022","journal-title":"Proc ICASSP"},{"article-title":"Towards better decoding and language model integration in sequence to sequence models","year":"2016","author":"chorowski","key":"ref31"},{"key":"ref30","first-page":"1195","article-title":"Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results","author":"tarvainen","year":"2017","journal-title":"Proc NeurIPS"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414594"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9052964"},{"key":"ref2","first-page":"577","article-title":"Attention-based models for speech recognition","author":"chorowski","year":"2015","journal-title":"Proc NeurIPS"},{"key":"ref1","first-page":"1764","article-title":"Towards end-to-end speech recognition with recurrent neural networks","author":"graves","year":"2014","journal-title":"Proc ICML"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2680"},{"article-title":"Krishna, kalpesh and toshniwal, shubham and livescu, karen","year":"2018","author":"krishna","key":"ref38"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414058"},{"key":"ref23","doi-asserted-by":"crossref","first-page":"741","DOI":"10.21437\/Interspeech.2021-740","article-title":"slim-IPL: Language-model-free iterative pseudo-labeling","author":"likhomanenko","year":"2021","journal-title":"Proc INTERSPEECH"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-571"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9054295"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1800"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1470"},{"key":"ref21","first-page":"2787","article-title":"Semi-supervised ASR by endto-end self-training","author":"chen","year":"2020","journal-title":"Proc INTERSPEECH"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746275"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU51503.2021.9688028"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2022.3195367"}],"event":{"name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","start":{"date-parts":[[2023,6,4]]},"location":"Rhodes Island, Greece","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":"https:\/\/summer-heart-0930.chufeiyun1688.workers.dev:443\/http\/xplorestaging.ieee.org\/ielx7\/10094559\/10094560\/10097071.pdf?arnumber=10097071","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,10,19]],"date-time":"2024-10-19T22:11:35Z","timestamp":1729375895000},"score":1,"resource":{"primary":{"URL":"https:\/\/summer-heart-0930.chufeiyun1688.workers.dev:443\/https\/ieeexplore.ieee.org\/document\/10097071\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,6,4]]},"references-count":49,"URL":"https:\/\/summer-heart-0930.chufeiyun1688.workers.dev:443\/https\/doi.org\/10.1109\/icassp49357.2023.10097071","relation":{},"subject":[],"published":{"date-parts":[[2023,6,4]]}}}