{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,6]],"date-time":"2025-12-06T16:48:53Z","timestamp":1765039733754,"version":"3.37.3"},"reference-count":32,"publisher":"IEEE","license":[{"start":{"date-parts":[[2024,4,14]],"date-time":"2024-04-14T00:00:00Z","timestamp":1713052800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,4,14]],"date-time":"2024-04-14T00:00:00Z","timestamp":1713052800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100007787","name":"National Research Centre","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100007787","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100007880","name":"Johns Hopkins University","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100007880","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100005801","name":"Facebook","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100005801","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006785","name":"Google","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006785","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100004318","name":"Microsoft","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100004318","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2024,4,14]]},"DOI":"10.1109\/icasspw62465.2024.10627008","type":"proceedings-article","created":{"date-parts":[[2024,8,15]],"date-time":"2024-08-15T17:19:18Z","timestamp":1723742358000},"page":"685-689","source":"Crossref","is-referenced-by-count":3,"title":["Training Early-Exit Architectures for Automatic Speech Recognition: Fine-Tuning Pre-Trained Models or Training from Scratch"],"prefix":"10.1109","author":[{"given":"George August","family":"Wright","sequence":"first","affiliation":[{"name":"University of Trento"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Umberto","family":"Cappellazzo","sequence":"additional","affiliation":[{"name":"University of Trento"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Salah","family":"Zaiem","sequence":"additional","affiliation":[{"name":"Institut Polytechnique de Paris,LTCI, Télécom Paris"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Desh","family":"Raj","sequence":"additional","affiliation":[{"name":"Johns Hopkins University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lucas Ondel","family":"Yang","sequence":"additional","affiliation":[{"name":"Universite Paris-Saclay,LISN, CNRS"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Daniele","family":"Falavigna","sequence":"additional","affiliation":[{"name":"Fondazione Bruno Kessler"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mohamed Nabih","family":"Ali","sequence":"additional","affiliation":[{"name":"Fondazione Bruno Kessler"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Alessio","family":"Brutti","sequence":"additional","affiliation":[{"name":"Fondazione Bruno Kessler"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3564242","article-title":"WH2D2N2: Distributed AI-enabled OK-ASN service for Web of Things","volume":"22","author":"Liang","year":"2023","journal-title":"ACM Transactions on Asian and Low-Resource Language Information Processing"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2020.2969775"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-021-01453-z"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6639201"},{"article-title":"Deep compression: Compressing deep neural network with pruning, trained quantization and Huffman coding","volume-title":"Proc. of 4th International Conference on Learning Representations (ICLR)","author":"Song","key":"ref5"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.521"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/icpr.2016.7900006"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00144"},{"key":"ref9","article-title":"HuBERT-EE: Early exiting HuBERT for efficient speech recognition"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/icassp49357.2023.10095335"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSPW59220.2023.10193042"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1007\/s12559-020-09734-4"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1145\/3587135.3592204"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00843"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3122291"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2022.3188113"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746863"},{"article-title":"BERT loses patience: Fast and robust inference with early exit","volume-title":"NIPS","author":"Zhou","key":"ref19"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.537"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.acl-main.204"},{"article-title":"Understanding and improving transformer from a multi-particle dynamic system point of view","year":"2019","author":"Lu","key":"ref22"},{"article-title":"Wav2Vec 2.0: A Framework for Self-Supervised Learning of Speech Representations","year":"2020","author":"Baevski","key":"ref23"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/1143844.1143891"},{"article-title":"Attention-based models for speech recognition","volume-title":"NIPS","author":"Chorowski","key":"ref25"},{"key":"ref26","doi-asserted-by":"crossref","DOI":"10.1109\/ICASSP.2017.7953075","article-title":"Joint CTC-attention based end-to-end speech recognition using multi-task learning","volume-title":"IEEE ICASSP","author":"Kim"},{"article-title":"SpeechBrain: A general-purpose speech toolkit","year":"2021","author":"Ravanelli","key":"ref27"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-99579-3_21"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-long.80"},{"key":"ref31","first-page":"1715","article-title":"Neural machine translation of rare words with subword units","volume-title":"Proc. of ACL","author":"R"},{"article-title":"RomeBERT: Robust training of multi-exit BERT","year":"2021","author":"Geng","key":"ref32"}],"event":{"name":"2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)","start":{"date-parts":[[2024,4,14]]},"location":"Seoul, Korea, Republic of","end":{"date-parts":[[2024,4,19]]}},"container-title":["2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10625769\/10625780\/10627008.pdf?arnumber=10627008","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,16]],"date-time":"2024-08-16T17:18:49Z","timestamp":1723828729000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10627008\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,4,14]]},"references-count":32,"URL":"https:\/\/doi.org\/10.1109\/icasspw62465.2024.10627008","relation":{},"subject":[],"published":{"date-parts":[[2024,4,14]]}}}