{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T19:45:01Z","timestamp":1776887101990,"version":"3.51.2"},"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"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2024,4,14]]},"DOI":"10.1109\/icassp48485.2024.10447448","type":"proceedings-article","created":{"date-parts":[[2024,3,18]],"date-time":"2024-03-18T18:56:31Z","timestamp":1710788191000},"page":"12056-12060","source":"Crossref","is-referenced-by-count":13,"title":["Retrieval Augmented End-to-End Spoken Dialog Models"],"prefix":"10.1109","author":[{"given":"Mingqiu","family":"Wang","sequence":"first","affiliation":[{"name":"Google DeepMind"}]},{"given":"Izhak","family":"Shafran","sequence":"additional","affiliation":[{"name":"Google DeepMind"}]},{"given":"Hagen","family":"Soltau","sequence":"additional","affiliation":[{"name":"Google DeepMind"}]},{"given":"Wei","family":"Han","sequence":"additional","affiliation":[{"name":"Google DeepMind"}]},{"given":"Yuan","family":"Cao","sequence":"additional","affiliation":[{"name":"Google DeepMind"}]},{"given":"Dian","family":"Yu","sequence":"additional","affiliation":[{"name":"Google DeepMind"}]},{"given":"Laurent","family":"El Shafey","sequence":"additional","affiliation":[{"name":"Google DeepMind"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU57964.2023.10389703"},{"key":"ref2","article-title":"Retrieval-augmented generation for knowledgeintensive nlp tasks","volume-title":"Proc. NeurIPS","author":"Lewis"},{"key":"ref3","article-title":"Retrieval augmented language model pre-training","volume-title":"Proc. ICML","author":"Guu"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-main.550"},{"key":"ref5","article-title":"Description-driven task-oriented dialog modeling","author":"Zhao","year":"2022"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-1037"},{"key":"ref7","article-title":"Multiwoz 2.1: Multi-domain dialogue state corrections and state tracking baselines","author":"Eric","year":"2019","journal-title":"CoRR"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.emnlp-main.669"},{"key":"ref9","article-title":"SLAM: A unified encoder for speech and language modeling via speech-text joint pre-training","author":"Bapna","year":"2021","journal-title":"CoRR"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2022-10937"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9746475"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747862"},{"key":"ref13","article-title":"Perceiver io: A general architecture for structured inputs & outputs","author":"Jaegle","year":"2021"},{"key":"ref14","article-title":"Flamingo: a visual language model for few-shot learning","author":"Alayrac","year":"2022","journal-title":"CoRR"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/taslp.2023.3288409"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096988"},{"key":"ref17","article-title":"Nearest neighbor machine translation","volume-title":"Proc. ICLR","author":"Khandelwal"},{"key":"ref18","article-title":"Improving language models by retrieving from trillions of tokens","author":"Borgeaud","year":"2021","journal-title":"CoRR"},{"key":"ref19","article-title":"Atlas: Few-shot learning with retrieval augmented language models","author":"Izacard","year":"2022"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1078"},{"key":"ref21","article-title":"Multi-domain dialogue state tracking as dynamic knowledge graph enhanced question answering","author":"Zhou","year":"2019","journal-title":"CoRR"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6394"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.emnlp-main.607"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.naacl-main.336"},{"key":"ref25","article-title":"Google usm: Scaling automatic speech recognition beyond 100 languages","author":"Zhang","year":"2023"},{"key":"ref26","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume-title":"JMLR","author":"Raffel"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/1143844.1143891"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10096065"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.5555\/3295222.3295349"},{"key":"ref30","article-title":"Accelerating large-scale inference with anisotropic vector quantization","volume-title":"Proc. ICML","author":"Guo"},{"key":"ref31","article-title":"Self-supervised learning with random-projection quantizer for speech recognition","volume-title":"Proc. ICML","author":"Chiu"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.21437\/interspeech.2020-3015"}],"event":{"name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Seoul, Korea, Republic of","start":{"date-parts":[[2024,4,14]]},"end":{"date-parts":[[2024,4,19]]}},"container-title":["ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/10445798\/10445803\/10447448.pdf?arnumber=10447448","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,2]],"date-time":"2024-08-02T04:42:44Z","timestamp":1722573764000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10447448\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,4,14]]},"references-count":32,"URL":"https:\/\/doi.org\/10.1109\/icassp48485.2024.10447448","relation":{},"subject":[],"published":{"date-parts":[[2024,4,14]]}}}