{"id":"https://openalex.org/W4375850714","doi":"https://doi.org/10.1109/icassp49357.2023.10095040","title":"Reducing the GAP Between Streaming and Non-Streaming Transducer-Based ASR by Adaptive Two-Stage Knowledge Distillation","display_name":"Reducing the GAP Between Streaming and Non-Streaming Transducer-Based ASR by Adaptive Two-Stage Knowledge Distillation","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4375850714","doi":"https://doi.org/10.1109/icassp49357.2023.10095040"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10095040","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10095040","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2306.15171","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5044202868","display_name":"Haitao Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Haitao Tang","raw_affiliation_strings":["iFLYTEK Research"],"affiliations":[{"raw_affiliation_string":"iFLYTEK Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014212007","display_name":"Yu Fu","orcid":"https://orcid.org/0000-0002-9795-7807"},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu Fu","raw_affiliation_strings":["Zhejiang University"],"affiliations":[{"raw_affiliation_string":"Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101926390","display_name":"Lei Sun","orcid":"https://orcid.org/0000-0001-7680-6455"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lei Sun","raw_affiliation_strings":["iFLYTEK Research"],"affiliations":[{"raw_affiliation_string":"iFLYTEK Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015690293","display_name":"Jiabin Xue","orcid":null},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiabin Xue","raw_affiliation_strings":["Harbin Institute of Technology"],"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100326831","display_name":"Dan Liu","orcid":"https://orcid.org/0000-0002-2198-3910"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dan Liu","raw_affiliation_strings":["iFLYTEK Research"],"affiliations":[{"raw_affiliation_string":"iFLYTEK Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101806000","display_name":"Yongchao Li","orcid":"https://orcid.org/0000-0002-4460-9702"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yongchao Li","raw_affiliation_strings":["iFLYTEK Research"],"affiliations":[{"raw_affiliation_string":"iFLYTEK Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053202704","display_name":"Zhiqiang Ma","orcid":"https://orcid.org/0000-0003-0006-2044"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhiqiang Ma","raw_affiliation_strings":["iFLYTEK Research"],"affiliations":[{"raw_affiliation_string":"iFLYTEK Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101925641","display_name":"Minghui Wu","orcid":"https://orcid.org/0000-0001-8179-7119"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Minghui Wu","raw_affiliation_strings":["iFLYTEK Research"],"affiliations":[{"raw_affiliation_string":"iFLYTEK Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101924451","display_name":"Jia Pan","orcid":"https://orcid.org/0000-0002-7073-1744"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jia Pan","raw_affiliation_strings":["iFLYTEK Research"],"affiliations":[{"raw_affiliation_string":"iFLYTEK Research","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011133553","display_name":"Genshun Wan","orcid":"https://orcid.org/0000-0002-5813-9430"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Genshun Wan","raw_affiliation_strings":["iFLYTEK Research"],"affiliations":[{"raw_affiliation_string":"iFLYTEK Research","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101201043","display_name":"Ming\u2019En Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ming\u2019En Zhao","raw_affiliation_strings":["iFLYTEK Research"],"affiliations":[{"raw_affiliation_string":"iFLYTEK Research","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":11,"corresponding_author_ids":["https://openalex.org/A5044202868"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.5274,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":{"value":0.70362234,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"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":0.9990000128746033,"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":0.9954000115394592,"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":0.7390173673629761},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6067497134208679},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.506620466709137},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.44697296619415283},{"id":"https://openalex.org/keywords/smoothness","display_name":"Smoothness","score":0.4181949496269226},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.39920124411582947},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3829771876335144},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.16557028889656067}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7390173673629761},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6067497134208679},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.506620466709137},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.44697296619415283},{"id":"https://openalex.org/C102634674","wikidata":"https://www.wikidata.org/wiki/Q868473","display_name":"Smoothness","level":2,"score":0.4181949496269226},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.39920124411582947},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3829771876335144},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.16557028889656067},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"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/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","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},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp49357.2023.10095040","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10095040","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2306.15171","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.15171","pdf_url":"https://arxiv.org/pdf/2306.15171","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2306.15171","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2306.15171","pdf_url":"https://arxiv.org/pdf/2306.15171","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4375850714.pdf"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1690739335","https://openalex.org/W1821462560","https://openalex.org/W1828163288","https://openalex.org/W2143612262","https://openalex.org/W2963414781","https://openalex.org/W3015194534","https://openalex.org/W3015412285","https://openalex.org/W3016010032","https://openalex.org/W3096518646","https://openalex.org/W3097777922","https://openalex.org/W3105966348","https://openalex.org/W3161873870","https://openalex.org/W3162649911","https://openalex.org/W3214456683","https://openalex.org/W4285242950","https://openalex.org/W4288088457","https://openalex.org/W4387385613","https://openalex.org/W6638523607","https://openalex.org/W6638749077","https://openalex.org/W6747398299","https://openalex.org/W6768086466","https://openalex.org/W6769806307","https://openalex.org/W6850390377"],"related_works":["https://openalex.org/W4388335561","https://openalex.org/W2970530566","https://openalex.org/W4288261899","https://openalex.org/W4307309205","https://openalex.org/W2967478618","https://openalex.org/W4385009901","https://openalex.org/W4385572700","https://openalex.org/W2393022482","https://openalex.org/W2377346130","https://openalex.org/W2997152889"],"abstract_inverted_index":{"Transducer":[0],"is":[1,12,34,55],"one":[2],"of":[3,66,87],"the":[4,17,59,64,67,72,96,114,145,150],"mainstream":[5],"frameworks":[6],"for":[7,144],"streaming":[8,18,152],"speech":[9],"recognition.":[10],"There":[11],"a":[13,119,141],"performance":[14],"gap":[15],"between":[16],"and":[19,40,91,140],"non-streaming":[20],"transducer":[21],"models":[22],"due":[23],"to":[24,35,57,126],"limited":[25],"context.":[26],"To":[27],"reduce":[28],"this":[29,76],"gap,":[30],"an":[31,80],"effective":[32],"way":[33],"ensure":[36,58],"that":[37],"their":[38],"hidden":[39,73,88,101],"output":[41,68,92,129],"distributions":[42],"are":[43],"consistent,":[44],"which":[45],"can":[46],"be":[47],"achieved":[48,132],"by":[49,106],"hierarchical":[50],"knowledge":[51,83],"distillation.":[52],"However,":[53],"it":[54],"difficult":[56],"distribution":[60,69],"consistency":[61],"simultaneously":[62],"because":[63],"learning":[65,90],"depends":[70],"on":[71],"one.":[74],"In":[75,95,113],"paper,":[77],"we":[78,99,117],"propose":[79],"adaptive":[81,123],"two-stage":[82],"distillation":[84],"method":[85,125],"consisting":[86],"layer":[89,93],"learning.":[94],"former":[97],"stage,":[98,116],"learn":[100,127],"representation":[102],"with":[103,149],"full":[104],"context":[105],"applying":[107],"mean":[108],"square":[109],"error":[110,138],"loss":[111],"function.":[112],"latter":[115],"design":[118],"power":[120],"transformation":[121],"based":[122],"smoothness":[124],"stable":[128],"distribution.":[130],"It":[131],"19%":[133],"relative":[134],"reduction":[135],"in":[136,154],"word":[137],"rate,":[139],"faster":[142],"response":[143],"first":[146],"token":[147],"compared":[148],"original":[151],"model":[153],"LibriSpeech":[155],"corpus.":[156]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1}],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2025-10-10T00:00:00"}