{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T06:59:55Z","timestamp":1775199595706,"version":"3.50.1"},"reference-count":46,"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"}],"funder":[{"DOI":"10.13039\/501100001321","name":"National Research Foundation","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100001321","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,12,6]]},"DOI":"10.1109\/asru65441.2025.11434599","type":"proceedings-article","created":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T19:48:04Z","timestamp":1775159284000},"page":"1-8","source":"Crossref","is-referenced-by-count":0,"title":["Diversity and complementarity of speech encoders across diverse tasks in a multi-modal large language model"],"prefix":"10.1109","author":[{"given":"Jeremy H. M.","family":"Wong","sequence":"first","affiliation":[{"name":"Institute of Infocomm Research (I2R), A* STAR,Singapore"}]},{"given":"Muhammad","family":"Huzaifah","sequence":"additional","affiliation":[{"name":"Institute of Infocomm Research (I2R), A* STAR,Singapore"}]},{"given":"Hardik B.","family":"Sailor","sequence":"additional","affiliation":[{"name":"Institute of Infocomm Research (I2R), A* STAR,Singapore"}]},{"given":"Shuo","family":"Sun","sequence":"additional","affiliation":[{"name":"Institute of Infocomm Research (I2R), A* STAR,Singapore"}]},{"given":"Kye Min","family":"Tan","sequence":"additional","affiliation":[{"name":"Institute of Infocomm Research (I2R), A* STAR,Singapore"}]},{"given":"Bin","family":"Wang","sequence":"additional","affiliation":[{"name":"Institute of Infocomm Research (I2R), A* STAR,Singapore"}]},{"given":"Qiongqiong","family":"Wang","sequence":"additional","affiliation":[{"name":"Institute of Infocomm Research (I2R), A* STAR,Singapore"}]},{"given":"Wenyu","family":"Zhang","sequence":"additional","affiliation":[{"name":"Institute of Infocomm Research (I2R), A* STAR,Singapore"}]},{"given":"Xunlong","family":"Zou","sequence":"additional","affiliation":[{"name":"Institute of Infocomm Research (I2R), A* STAR,Singapore"}]},{"given":"Nancy F.","family":"Chen","sequence":"additional","affiliation":[{"name":"Institute of Infocomm Research (I2R), A* STAR,Singapore"}]},{"given":"Ai Ti","family":"Aw","sequence":"additional","affiliation":[{"name":"Institute of Infocomm Research (I2R), A* STAR,Singapore"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Language models are unsupervised multitask learners","author":"Radford","year":"2019"},{"key":"ref2","article-title":"Language models are few-shot learners","author":"Brown","year":"2020","journal-title":"NeurIPS, Vancouver, Canada"},{"key":"ref3","article-title":"Retrieval-augmented generation for knowledge-intensive NLP tasks","volume-title":"NeurIPS","author":"Lewis"},{"key":"ref4","first-page":"19730","article-title":"BLIP-2: bootstrapping languageimage pre-training with frozen image encoders and large language models","volume-title":"ICML","author":"Li"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-long.393"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-emnlp.263"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/taslpro.2025.3648793"},{"key":"ref8","article-title":"SALMONN: towards generic hearing abilities for large language models","volume-title":"ICLR","author":"Tang"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49660.2025.10888128"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3122291"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2022.3188113"},{"key":"ref12","article-title":"Seamless: multilingual expressive and streaming speech translation","volume-title":"Tech. Rep."},{"key":"ref13","article-title":"MERaLiONSpeechEncoder: towards a speech foundation model for Singapore and beyond","volume-title":"Tech. Rep., Institute for Infocomm Research","author":"Huzaifah","year":"2024"},{"key":"ref14","first-page":"28492","article-title":"Robust speech recognition via large-scale weak supervision","volume-title":"ICML","author":"Radford"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2024-1194"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-acl.931"},{"key":"ref17","first-page":"5178","article-title":"BEATs: audio pre-training with acoustic tokenizers","volume-title":"ICML","author":"Chen"},{"key":"ref18","article-title":"MERT: acoustic music understanding model with large-scale self-supervised training","volume-title":"ICLR","author":"Li"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10448257"},{"key":"ref20","article-title":"MERaLiON-AudioLLM: bridging audio and language with large language models","volume-title":"Tech. Rep.","author":"He","year":"2025"},{"key":"ref21","article-title":"Gemma 2: improving open language models at a practical size","volume-title":"Tech. Rep., Google DeepMind","year":"2024"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2017.12.012"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-1525"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-1965"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICSDA.2017.8384449"},{"key":"ref27","first-page":"4218","article-title":"Common Voice: a massively-multilingual speech corpus","volume-title":"LREC","author":"Ardila"},{"key":"ref28","article-title":"The People\u2019s Speech: a large-scale diverse English speech recognition dataset for commercial usage","author":"Galvez"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-2027"},{"key":"ref30","first-page":"119","article-title":"AudioCaps: generating captions for audios in the wild","volume-title":"NAACL-HLT","author":"Kim"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2024.3419446"},{"key":"ref32","article-title":"Instruction tuning with GPT-4","author":"Peng","year":"2023","journal-title":"arXiv preprint arXiv:2304.03277"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00264"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.496"},{"key":"ref35","first-page":"3459","article-title":"Spoken SQuAD: a study of mitigating the impact of speech recognition errors on listening comprehension","author":"Li","year":"2018","journal-title":"Interspeech, Hyderabad, India"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1007\/s10579-008-9076-6"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1050"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1016\/j.csl.2019.101027"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9052990"},{"key":"ref40","article-title":"Decoupled weight decay regularization","volume-title":"ICLR","author":"Loshchilov"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.naacl-long.218"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2021-1775"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9052942"},{"key":"ref44","article-title":"Google USM: scaling automatic speech recognition beyond 100 languages","author":"Zhang","year":"2023","journal-title":"arXiv preprint arXiv:2303.01037"},{"key":"ref45","first-page":"3915","article-title":"Self-supervised learning with random-projection quantizer for speech recognition","author":"Chiu","year":"2022","journal-title":"ICML, Baltimore, USA"},{"key":"ref46","article-title":"Bootstrap your own latent: a new approach to self-supervised learning","author":"Grill","year":"2020","journal-title":"NeurIPS, Vancouver, Canada"}],"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\/11434599.pdf?arnumber=11434599","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T04:56:38Z","timestamp":1775192198000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11434599\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,6]]},"references-count":46,"URL":"https:\/\/doi.org\/10.1109\/asru65441.2025.11434599","relation":{},"subject":[],"published":{"date-parts":[[2025,12,6]]}}}