{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T06:23:59Z","timestamp":1774419839512,"version":"3.50.1"},"reference-count":26,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,4,6]],"date-time":"2025-04-06T00:00:00Z","timestamp":1743897600000},"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":[[2025,4,6]]},"DOI":"10.1109\/icassp49660.2025.10890328","type":"proceedings-article","created":{"date-parts":[[2025,3,12]],"date-time":"2025-03-12T13:52:43Z","timestamp":1741787563000},"page":"1-5","source":"Crossref","is-referenced-by-count":2,"title":["Faster Speech-LLaMA Inference with Multi-token Prediction"],"prefix":"10.1109","author":[{"given":"Desh","family":"Raj","sequence":"first","affiliation":[{"name":"Meta,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gil","family":"Keren","sequence":"additional","affiliation":[{"name":"Meta,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Junteng","family":"Jia","sequence":"additional","affiliation":[{"name":"Meta,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jay","family":"Mahadeokar","sequence":"additional","affiliation":[{"name":"Meta,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ozlem","family":"Kalinli","sequence":"additional","affiliation":[{"name":"Meta,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10447605"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.naacl-long.309"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10445874"},{"key":"ref4","article-title":"Salmonn: Towards generic hearing abilities for large language models","volume-title":"ICLR","author":"Tang"},{"key":"ref5","article-title":"Training speech recognition models to follow instructions","volume-title":"Workshop on Instruction Tuning and Instruction Following at NeurIPS","author":"Lai"},{"key":"ref6","article-title":"Sequence transduction with recurrent neural networks","volume-title":"ICML Representation Learning Workshop","author":"Graves"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/icassp.2016.7472621"},{"key":"ref8","article-title":"Efficiently scaling transformer inference","volume-title":"Machine Learning and Systems","author":"Pope"},{"key":"ref9","article-title":"Parallel wavenet: Fast high-fidelity speech synthesis","volume-title":"ICML","author":"van den Oord"},{"key":"ref10","article-title":"Fast transformer decoding: One write-head is all you need","author":"Shazeer","year":"2019"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.298"},{"key":"ref12","article-title":"Efficient neural audio synthesis","volume-title":"ICML","author":"Kalchbrenner"},{"key":"ref13","article-title":"LLM.int8(): 8-bit matrix multiplication for transformers at scale","volume-title":"NeurIPS","author":"Dettmers"},{"key":"ref14","article-title":"Smoothquant: Accurate and efficient post-training quantization for large language models","volume-title":"ICML","author":"Xiao"},{"key":"ref15","article-title":"Fast inference from transformers via speculative decoding","volume-title":"ICML","author":"Leviathan"},{"key":"ref16","article-title":"Accelerating large language model decoding with speculative sampling","author":"Chen","year":"2023"},{"key":"ref17","article-title":"Better & faster large language models via multi-token prediction","volume-title":"ICML","author":"Gloeckle"},{"key":"ref18","article-title":"Medusa: Simple llm inference acceleration framework with multiple decoding heads","volume-title":"ICML","author":"Cai"},{"key":"ref19","article-title":"Speculative streaming: Fast llm inference without auxiliary models","author":"Bhendawade","year":"2024"},{"key":"ref20","article-title":"Blockwise parallel decoding for deep autoregressive models","volume-title":"NeurIPS","author":"Stern"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/icassp48485.2024.10448204"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/icassp48485.2024.10447553"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2015.7178964"},{"key":"ref24","article-title":"Common voice: A massively-multilingual speech corpus","volume-title":"LREC","author":"Ardila"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-long.80"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-2826"}],"event":{"name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","location":"Hyderabad, India","start":{"date-parts":[[2025,4,6]]},"end":{"date-parts":[[2025,4,11]]}},"container-title":["ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10887540\/10887541\/10890328.pdf?arnumber=10890328","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T05:24:58Z","timestamp":1774416298000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10890328\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,6]]},"references-count":26,"URL":"https:\/\/doi.org\/10.1109\/icassp49660.2025.10890328","relation":{},"subject":[],"published":{"date-parts":[[2025,4,6]]}}}