{"id":"https://openalex.org/W3160628828","doi":"https://doi.org/10.1109/icassp39728.2021.9413692","title":"Improving Streaming Automatic Speech Recognition with Non-Streaming Model Distillation on Unsupervised Data","display_name":"Improving Streaming Automatic Speech Recognition with Non-Streaming Model Distillation on Unsupervised Data","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3160628828","doi":"https://doi.org/10.1109/icassp39728.2021.9413692","mag":"3160628828"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9413692","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9413692","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5017597250","display_name":"Thibault Doutre","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Thibault Doutre","raw_affiliation_strings":["Google Inc, USA"],"affiliations":[{"raw_affiliation_string":"Google Inc, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100750907","display_name":"Wei Han","orcid":"https://orcid.org/0000-0002-4201-9645"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wei Han","raw_affiliation_strings":["Google Inc, USA"],"affiliations":[{"raw_affiliation_string":"Google Inc, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101706633","display_name":"Min Ma","orcid":"https://orcid.org/0000-0002-3132-4898"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Min Ma","raw_affiliation_strings":["Google Inc, USA"],"affiliations":[{"raw_affiliation_string":"Google Inc, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089287882","display_name":"Zhiyun Lu","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhiyun Lu","raw_affiliation_strings":["Google Inc, USA"],"affiliations":[{"raw_affiliation_string":"Google Inc, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027763497","display_name":"Chung\u2010Cheng Chiu","orcid":"https://orcid.org/0000-0001-9729-4778"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chung-Cheng Chiu","raw_affiliation_strings":["Google Inc, USA"],"affiliations":[{"raw_affiliation_string":"Google Inc, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112901893","display_name":"Ruoming Pang","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ruoming Pang","raw_affiliation_strings":["Google Inc, USA"],"affiliations":[{"raw_affiliation_string":"Google Inc, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000078382","display_name":"Arun Narayanan","orcid":"https://orcid.org/0009-0008-3325-8928"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Arun Narayanan","raw_affiliation_strings":["Google Inc, USA"],"affiliations":[{"raw_affiliation_string":"Google Inc, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091715883","display_name":"Ananya Misra","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ananya Misra","raw_affiliation_strings":["Google Inc, USA"],"affiliations":[{"raw_affiliation_string":"Google Inc, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100433648","display_name":"Yu Zhang","orcid":"https://orcid.org/0000-0002-9505-1833"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yu Zhang","raw_affiliation_strings":["Google Inc, USA"],"affiliations":[{"raw_affiliation_string":"Google Inc, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5103187717","display_name":"Liangliang Cao","orcid":"https://orcid.org/0000-0003-0900-1512"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Liangliang Cao","raw_affiliation_strings":["Google Inc, USA"],"affiliations":[{"raw_affiliation_string":"Google Inc, USA","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":10,"corresponding_author_ids":["https://openalex.org/A5017597250"],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":1.9035,"has_fulltext":false,"cited_by_count":15,"citation_normalized_percentile":{"value":0.88118339,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"6558","last_page":"6562"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998999834060669,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9998000264167786,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9994999766349792,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8474932312965393},{"id":"https://openalex.org/keywords/streaming-data","display_name":"Streaming data","score":0.6268525719642639},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5645627379417419},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5341542959213257},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.5327638983726501},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.48739272356033325},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4848578870296478},{"id":"https://openalex.org/keywords/streaming-algorithm","display_name":"Streaming algorithm","score":0.4533698558807373},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.44889217615127563},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.42090386152267456},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.41656067967414856},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.4132387340068817},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3710874617099762},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.18573147058486938},{"id":"https://openalex.org/keywords/database","display_name":"Database","score":0.11472651362419128}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8474932312965393},{"id":"https://openalex.org/C2777611316","wikidata":"https://www.wikidata.org/wiki/Q39045282","display_name":"Streaming data","level":2,"score":0.6268525719642639},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5645627379417419},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5341542959213257},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.5327638983726501},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.48739272356033325},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4848578870296478},{"id":"https://openalex.org/C187166803","wikidata":"https://www.wikidata.org/wiki/Q2835831","display_name":"Streaming algorithm","level":3,"score":0.4533698558807373},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.44889217615127563},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.42090386152267456},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41656067967414856},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.4132387340068817},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3710874617099762},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.18573147058486938},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.11472651362419128},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C127313418","wikidata":"https://www.wikidata.org/wiki/Q1069","display_name":"Geology","level":0,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C111368507","wikidata":"https://www.wikidata.org/wiki/Q43518","display_name":"Oceanography","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9413692","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9413692","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","score":0.7099999785423279,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":47,"referenced_works":["https://openalex.org/W854541894","https://openalex.org/W1494198834","https://openalex.org/W1828163288","https://openalex.org/W2033256038","https://openalex.org/W2121879602","https://openalex.org/W2327501763","https://openalex.org/W2592335154","https://openalex.org/W2889129739","https://openalex.org/W2936123380","https://openalex.org/W2936774411","https://openalex.org/W2962760690","https://openalex.org/W2973122799","https://openalex.org/W2976223659","https://openalex.org/W2981696585","https://openalex.org/W2982413405","https://openalex.org/W2995181338","https://openalex.org/W2995746049","https://openalex.org/W3004728855","https://openalex.org/W3008181812","https://openalex.org/W3015194534","https://openalex.org/W3015927303","https://openalex.org/W3015974384","https://openalex.org/W3016010032","https://openalex.org/W3025165719","https://openalex.org/W3026041220","https://openalex.org/W3030437843","https://openalex.org/W3035160371","https://openalex.org/W3037029661","https://openalex.org/W3096518646","https://openalex.org/W3096702180","https://openalex.org/W3097777922","https://openalex.org/W3125815078","https://openalex.org/W3128478537","https://openalex.org/W3149629662","https://openalex.org/W4288088457","https://openalex.org/W6623517193","https://openalex.org/W6629717138","https://openalex.org/W6638749077","https://openalex.org/W6768222176","https://openalex.org/W6769647859","https://openalex.org/W6769806307","https://openalex.org/W6771467084","https://openalex.org/W6771812881","https://openalex.org/W6775998690","https://openalex.org/W6776544788","https://openalex.org/W6779924452","https://openalex.org/W6790121257"],"related_works":["https://openalex.org/W3191101401","https://openalex.org/W2241126753","https://openalex.org/W1824789409","https://openalex.org/W2991654599","https://openalex.org/W201138236","https://openalex.org/W3122033752","https://openalex.org/W2128692898","https://openalex.org/W2398953209","https://openalex.org/W1733902306","https://openalex.org/W3002138439"],"abstract_inverted_index":{"Streaming":[0],"end-to-end":[1],"automatic":[2],"speech":[3,23],"recognition":[4],"(ASR)":[5],"models":[6,18,44,96,121],"are":[7,19,28,139],"widely":[8],"used":[9,80],"on":[10,71,124,128,160],"smart":[11],"speakers":[12],"and":[13,55],"on-device":[14],"applications.":[15],"Since":[16],"these":[17],"expected":[20],"to":[21,30,38,68,81,97,99,141,148],"transcribe":[22],"with":[24,33],"minimal":[25],"latency,":[26],"they":[27],"constrained":[29],"be":[31],"causal":[32],"no":[34],"future":[35],"context,":[36],"compared":[37],"their":[39],"non-streaming":[40,49,62,156],"counterparts.":[41],"Consequently,":[42],"streaming":[43,85,95,151],"usually":[45],"perform":[46],"worse":[47],"than":[48],"models.":[50,87],"We":[51],"propose":[52],"a":[53,61,66,149,155],"novel":[54],"effective":[56],"learning":[57],"method":[58],"by":[59,145,153],"leveraging":[60,154],"ASR":[63,86],"model":[64,152,158],"as":[65,167],"teacher":[67,157],"generate":[69],"transcripts":[70],"an":[72],"arbitrarily":[73],"large":[74],"data":[75,130,166],"set,":[76],"which":[77],"is":[78],"then":[79],"distill":[82],"knowledge":[83],"into":[84],"This":[88],"way,":[89],"we":[90,138],"scale":[91],"the":[92,114,143,161,168],"training":[93],"of":[94,103,119,164],"up":[98],"3":[100],"million":[101],"hours":[102],"YouTube":[104,129],"audio.":[105],"Experiments":[106],"show":[107],"that":[108],"our":[109],"approach":[110],"can":[111],"significantly":[112],"reduce":[113,142],"word":[115],"error":[116],"rate":[117],"(WER)":[118],"RNN-T":[120],"not":[122],"only":[123],"LibriSpeech":[125],"but":[126],"also":[127],"in":[131,136],"four":[132],"languages.":[133],"For":[134],"example,":[135],"French,":[137],"able":[140],"WER":[144],"16.4%":[146],"relatively":[147],"baseline":[150],"trained":[159],"same":[162],"amount":[163],"labeled":[165],"baseline.":[169]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":6},{"year":2021,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}