{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T12:05:22Z","timestamp":1791893122738,"version":"4.3.4"},"reference-count":81,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"12","license":[{"start":{"date-parts":[[2017,12,1]],"date-time":"2017-12-01T00:00:00Z","timestamp":1512086400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE\/ACM Trans. Audio Speech Lang. Process."],"published-print":{"date-parts":[[2017,12]]},"DOI":"10.1109\/taslp.2017.2756440","type":"journal-article","created":{"date-parts":[[2017,9,25]],"date-time":"2017-09-25T18:13:03Z","timestamp":1506363183000},"page":"2410-2423","source":"Crossref","is-referenced-by-count":200,"title":["Toward Human Parity in Conversational Speech Recognition"],"prefix":"10.1109","volume":"25","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2854-1143","authenticated-orcid":false,"given":"Wayne","family":"Xiong","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jasha","family":"Droppo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4550-7908","authenticated-orcid":false,"given":"Xuedong","family":"Huang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Frank","family":"Seide","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3474-2451","authenticated-orcid":false,"given":"Michael L.","family":"Seltzer","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9925-905X","authenticated-orcid":false,"given":"Andreas","family":"Stolcke","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0520-6844","authenticated-orcid":false,"given":"Dong","family":"Yu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Geoffrey","family":"Zweig","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2015.2400218"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472719"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/E17-2025"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1145\/2505515.2505665"},{"key":"ref76","first-page":"1058","article-title":"1-bit stochastic gradient descent and its application to data-parallel\n distributed training of speech DNNs","author":"seide","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref77","article-title":"Adam: A method for stochastic optimization","author":"kingma","year":"0","journal-title":"Proc Int Conf Learn Represent"},{"key":"ref74","year":"2016"},{"key":"ref39","first-page":"746","article-title":"Linguistic\n regularities in continuous space word representations","volume":"13","author":"mikolov","year":"2013","journal-title":"HLT-NAACL"},{"key":"ref75","article-title":"An introduction to computational networks and the\n Computational Network Toolkit","author":"yu","year":"2014"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1007\/3-540-33486-6_6"},{"key":"ref78","first-page":"524","article-title":"Insights into spoken language gleaned from phonetic transcription of the Switchboard corpus","author":"greenberg","year":"0","journal-title":"Proc Int Conf Spoken Lang Process"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2013.6639347"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2016.2602884"},{"key":"ref32","first-page":"3259","article-title":"Very deep convolutional neural\n networks for LVCSR","author":"bi","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472620"},{"key":"ref30","first-page":"3140","article-title":"The IBM 2015 English conversational\n telephone speech recognition system","author":"saon","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-473"},{"key":"ref36","first-page":"194","article-title":"LSTM neural networks for language modeling","author":"sundermeyer","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/SLT.2012.6424228"},{"key":"ref34","first-page":"1045","article-title":"Recurrent neural network based language model","author":"mikolov","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2013.6707705"},{"key":"ref62","first-page":"441","article-title":"Sequential classification criteria for NNs in automatic speech recognition","author":"wang","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-1460"},{"key":"ref63","first-page":"2345","article-title":"Sequence-discriminative training of deep\n neural networks","author":"vesely","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref28","first-page":"338","article-title":"Long short-term memory\n recurrent neural network architectures for large scale acoustic modeling","author":"sak","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-595"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472694"},{"key":"ref66","first-page":"901","article-title":"SRILM—An extensible language modeling toolkit","author":"stolcke","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref29","first-page":"1468","article-title":"Fast and accurate recurrent neural network acoustic\n models for speech recognition","author":"sak","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref67","first-page":"1820","article-title":"Efficient estimation of maximum entropy language models with N-gram features: An SRILM extension","author":"alum\u00e4e","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472829"},{"key":"ref69","first-page":"297","article-title":"Noise-contrastive estimation: A new estimation principle for unnormalized\n statistical models","volume":"9","author":"gutmann","year":"0","journal-title":"Proc 13th Int Conf Artif Intell Statist"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1038\/nature16961"},{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/S0004-3702(01)00129-1"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2017.7953159"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1989.1.2.270"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.59.2229"},{"key":"ref24","first-page":"255","article-title":"Convolutional networks for images, speech, and time series","author":"lecun","year":"1995","journal-title":"The Handbook of Brain Theory and Neural Networks"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/29.21701"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1016\/0885-2308(91)90010-N"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1989.1.4.541"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref51","article-title":"Highway networks","author":"srivastava","year":"2015","journal-title":"CoRR"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2010.2064307"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-580"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1126\/science.1245994"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2005.06.042"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1989.266376"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2016-251"},{"key":"ref53","first-page":"448","article-title":"Batch normalization: Accelerating deep network training by reducing internal covariate shift","volume":"37","author":"ioffe","year":"0","journal-title":"Proceedings of the 32nd Intl Conf on Machine Learning"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2016.7472646"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1992.225858"},{"key":"ref11","first-page":"69","article-title":"The Fisher\n corpus: A resource for the next generations of speech-to-text","volume":"4","author":"cieri","year":"0","journal-title":"Proc 2nd Int Conf Lang Res Eval"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.1997.659110"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2006.879814"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2006.878257"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/TASL.2006.879807"},{"key":"ref15","article-title":"The AT&T 2001 LVCSR system","author":"ljolje","year":"0","journal-title":"Proc NIST LVCSR Workshop"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2003.1198755"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-1695"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2004.1325969"},{"key":"ref18","first-page":"437","article-title":"Conversational\n speech transcription using context-dependent deep neural networks","author":"seide","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1016\/S0167-6393(97)00021-6"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2012.6288864"},{"key":"ref4","first-page":"97?100","article-title":"Super-human multi-talker speech recognition: The IBM 2006 speech\n separation challenge system","author":"kristjansson","year":"0","journal-title":"Proc INTERSPEECH"},{"key":"ref3","first-page":"173","article-title":"Deep Speech 2: End-to-end speech recognition\n in English and Mandarin","volume":"48","author":"amodei","year":"0","journal-title":"Proc 33rd Int Conf Mach Learn"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/ASRU.2003.1318488"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2014.6854681"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.3115\/1075527.1075614"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.1988.196669"},{"key":"ref49","article-title":"Very deep convolutional networks for large-scale image recognition","author":"simonyan","year":"2014"},{"key":"ref9","first-page":"11","article-title":"The 1996 broadcast news speech and\n language-model corpus","author":"graff","year":"0","journal-title":"Proc DARPA Spoken Language Systems Technology Workshop"},{"key":"ref46","article-title":"The Microsoft 2017 conversational speech recognition system","author":"xiong","year":"2017"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-1544"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-405"},{"key":"ref47","first-page":"2915","article-title":"Transcription methods for consistency, volume and efficiency","author":"glenn","year":"0","journal-title":"Proc 2nd Int Conf Lang Res Eval"},{"key":"ref42","article-title":"The SRI March 2000 Hub-5 conversational speech\n transcription system","author":"stolcke","year":"0","journal-title":"Proc NIST Speech Transcription Workshop"},{"key":"ref41","article-title":"Posterior probability decoding, confidence estimation, and system combination","author":"evermann","year":"0","journal-title":"Proc NIST Speech Transcription Workshop"},{"key":"ref44","article-title":"Deep speech: Scaling up end-to-end speech\n recognition","author":"hannun","year":"2014"},{"key":"ref43","first-page":"3104","article-title":"Sequence to\n sequence learning with neural networks","author":"sutskever","year":"0","journal-title":"Proc Adv Neural Inf Process Syst"}],"container-title":["IEEE\/ACM Transactions on Audio, Speech, and Language Processing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/6570655\/8114348\/08049322.pdf?arnumber=8049322","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,1,12]],"date-time":"2022-01-12T16:24:07Z","timestamp":1642004647000},"score":1,"resource":{"primary":{"URL":"http:\/\/ieeexplore.ieee.org\/document\/8049322\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2017,12]]},"references-count":81,"journal-issue":{"issue":"12"},"URL":"https:\/\/doi.org\/10.1109\/taslp.2017.2756440","relation":{},"ISSN":["2329-9290","2329-9304"],"issn-type":[{"value":"2329-9290","type":"print"},{"value":"2329-9304","type":"electronic"}],"subject":[],"published":{"date-parts":[[2017,12]]}}}