{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,5]],"date-time":"2026-06-05T15:17:15Z","timestamp":1780672635914,"version":"3.54.1"},"reference-count":38,"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:\/\/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:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2023,6,4]]},"DOI":"10.1109\/icassp49357.2023.10094796","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":15,"title":["Massively Multilingual Shallow Fusion with Large Language Models"],"prefix":"10.1109","author":[{"given":"Ke","family":"Hu","sequence":"first","affiliation":[{"name":"Google LLC,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tara N.","family":"Sainath","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bo","family":"Li","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nan","family":"Du","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yanping","family":"Huang","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrew M.","family":"Dai","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yu","family":"Zhang","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rodrigo","family":"Cabrera","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhifeng","family":"Chen","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Trevor","family":"Strohman","sequence":"additional","affiliation":[{"name":"Google LLC,USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682336"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7953075"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU46091.2019.9003906"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1855"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2017-343"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462682"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU51503.2021.9688018"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8683307"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2017.8268950"},{"key":"ref10","author":"Bapna","year":"2021","journal-title":"SLAM: A unified encoder for speech and language modeling via speech-text joint pre-training"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-long.105"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747555"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10937"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/SLT54892.2023.10022774"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2021-206"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-2075"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10820"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8682918"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2225"},{"key":"ref20","first-page":"1877","article-title":"Language models are few-shot learners","volume":"33","author":"Brown","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref21","author":"Wang","year":"2021","journal-title":"GPT-J-6B: A 6 billion parameter autoregressive language model"},{"key":"ref22","author":"Adiwardana","year":"2020","journal-title":"Towards a human-like open-domain chatbot"},{"key":"ref23","author":"Thoppilan","year":"2022","journal-title":"LaMDA: Language models for dialog applications"},{"key":"ref24","author":"Smith","year":"2022","journal-title":"Using DeepSpeed and Megatron to train Megatron-Turing NLG 530B, a large-scale generative language model"},{"key":"ref25","first-page":"5547","article-title":"GLaM: Efficient scaling of language models with mixture-of-experts","volume-title":"International Conference on Machine Learning","author":"Du"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/jstsp.2022.3182537"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2018.8462682"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2017-343"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2021-206"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-3015"},{"key":"ref31","author":"Shazeer","year":"2017","journal-title":"Outrageously large neural networks: The sparsely-gated mixture-of-experts layer"},{"key":"ref32","author":"Lepikhin","year":"2020","journal-title":"GShard: Scaling giant models with conditional computation and automatic sharding"},{"key":"ref33","author":"Xu","year":"2021","journal-title":"GSPMD: General and scalable parallelization for ML computation graphs"},{"key":"ref34","first-page":"4596","article-title":"Adafactor: Adaptive learning rates with sublinear memory cost","volume-title":"International Conference on Machine Learning","author":"Shazeer"},{"key":"ref35","author":"Aarohi","year":"2022","journal-title":"Beyond the imitation game: Quantifying and extrapolating the capabilities of language models"},{"key":"ref36","volume-title":"Artificial Intelligence at Google: Our Principles"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-1465"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-243"}],"event":{"name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Rhodes Island, Greece","start":{"date-parts":[[2023,6,4]]},"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":"http:\/\/xplorestaging.ieee.org\/ielx7\/10094559\/10094560\/10094796.pdf?arnumber=10094796","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,2,29]],"date-time":"2024-02-29T22:33:40Z","timestamp":1709246020000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10094796\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,6,4]]},"references-count":38,"URL":"https:\/\/doi.org\/10.1109\/icassp49357.2023.10094796","relation":{},"subject":[],"published":{"date-parts":[[2023,6,4]]}}}