{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,12]],"date-time":"2025-10-12T04:57:32Z","timestamp":1760245052786,"version":"3.28.0"},"reference-count":32,"publisher":"IEEE","license":[{"start":{"date-parts":[[2019,5,1]],"date-time":"2019-05-01T00:00:00Z","timestamp":1556668800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2019,5,1]],"date-time":"2019-05-01T00:00:00Z","timestamp":1556668800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2019,5,1]],"date-time":"2019-05-01T00:00:00Z","timestamp":1556668800000},"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":[[2019,5]]},"DOI":"10.1109\/icassp.2019.8682888","type":"proceedings-article","created":{"date-parts":[[2019,4,17]],"date-time":"2019-04-17T16:01:56Z","timestamp":1555516916000},"page":"5706-5710","source":"Crossref","is-referenced-by-count":17,"title":["Distributed Deep Learning Strategies for Automatic Speech Recognition"],"prefix":"10.1109","author":[{"given":"Wei","family":"Zhang","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaodong","family":"Cui","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ulrich","family":"Finkler","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Brian","family":"Kingsbury","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"George","family":"Saon","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"David","family":"Kung","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Michael","family":"Picheny","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ICDM.2017.161"},{"key":"ref31","article-title":"Adacomp : Adaptive residual gradient compression for data-parallel distributed training","author":"chen","year":"2017","journal-title":"AAAI"},{"key":"ref30","article-title":"Training deep neural networks with 8-bit floating point numbers","author":"wang","year":"2018","journal-title":"NIPS"},{"key":"ref10","first-page":"173","article-title":"Deep speech 2 : End-to-end speech recognition in english and mandarin","author":"amodei","year":"2016","journal-title":"ICML’16"},{"key":"ref11","article-title":"Asynchronous decentralized parallel stochastic gradient descent","author":"lian","year":"2018","journal-title":"ICML"},{"key":"ref12","article-title":"Large scale distributed deep networks","author":"dean","year":"2012","journal-title":"NIPS"},{"key":"ref13","first-page":"2350","article-title":"Staleness-aware async-sgd for distributed deep learning","author":"zhang","year":"2016","journal-title":"Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence IJCAI 2016"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpdc.2008.09.002"},{"journal-title":"Effectively Scaling Deep Learning Frameworks","year":"0","key":"ref15"},{"journal-title":"NCCL Optimized primitives for collective multi-GPU communication","year":"0","key":"ref16"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/1327452.1327492"},{"key":"ref18","article-title":"Revisiting distributed synchronous sgd","author":"chen","year":"2016","journal-title":"International Conference on Learning Representations (ICLR) Workshop Track"},{"key":"ref19","article-title":"Model accuracy and runtime tradeoff in distributed deep learning: A systematic study","author":"zhang","year":"2016","journal-title":"IEEE International Conference on Data Mining"},{"journal-title":"The nvlink fabric","year":"0","key":"ref28"},{"key":"ref4","article-title":"Accurate, large minibatch SGD: training imagenet in 1 hour","volume":"abs 1706 2677","author":"goyal","year":"2017","journal-title":"CoRR"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/3075564.3075579"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2017.2756440"},{"key":"ref6","article-title":"Scaling SGD batch size to 32k for imagenet training","volume":"abs 1708 3888","author":"you","year":"2017","journal-title":"CoRR"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472805"},{"key":"ref5","article-title":"Powerai DDL","volume":"abs 1708 2188","author":"cho","year":"2017","journal-title":"CoRR"},{"key":"ref8","first-page":"1509","article-title":"Terngrad: Ternary gradients to reduce communication in distributed deep learning","author":"wen","year":"2017","journal-title":"NIPS’01"},{"key":"ref7","article-title":"Highly scalable deep learning training system with mixed-precision: Training imagenet in four minutes","volume":"abs 1807 11205","author":"jia","year":"2018","journal-title":"CoRR"},{"key":"ref2","doi-asserted-by":"crossref","DOI":"10.21437\/Interspeech.2017-405","article-title":"English conversational telephone speech recognition by humans and machines","author":"saon","year":"2017","journal-title":"InterSpeech"},{"key":"ref9","doi-asserted-by":"crossref","DOI":"10.21437\/Interspeech.2014-274","article-title":"1-bit stochastic gradient descent and application to data-parallel distributed training of speech dnns","author":"seide","year":"2014","journal-title":"Proc INTERSPEECH 2014"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2012.2205597"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CAHPC.2018.8645935"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1147\/JRD.2017.2709198"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-6301"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-30218-6_19"},{"key":"ref23","article-title":"Can decentralized algorithms outperform centralized algorithmsƒ A case study for decentralized parallel stochastic gradient descent","author":"lian","year":"2017","journal-title":"NIPS"},{"journal-title":"Operating Systems Three Easy Pieces","year":"2015","author":"arpaci-dusseau","key":"ref26"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/358818.358824"}],"event":{"name":"ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","start":{"date-parts":[[2019,5,12]]},"location":"Brighton, United Kingdom","end":{"date-parts":[[2019,5,17]]}},"container-title":["ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/8671773\/8682151\/08682888.pdf?arnumber=8682888","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,9,16]],"date-time":"2022-09-16T01:35:55Z","timestamp":1663292155000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/8682888\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,5]]},"references-count":32,"URL":"https:\/\/doi.org\/10.1109\/icassp.2019.8682888","relation":{},"subject":[],"published":{"date-parts":[[2019,5]]}}}