{"id":"https://openalex.org/W3184976814","doi":"https://doi.org/10.21437/interspeech.2021-2075","title":"Minimum Word Error Rate Training with Language Model Fusion for End-to-End Speech Recognition","display_name":"Minimum Word Error Rate Training with Language Model Fusion for End-to-End Speech Recognition","publication_year":2021,"publication_date":"2021-08-27","ids":{"openalex":"https://openalex.org/W3184976814","doi":"https://doi.org/10.21437/interspeech.2021-2075","mag":"3184976814"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2021-2075","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-2075","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2021","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/A5101749753","display_name":"Zhong Meng","orcid":"https://orcid.org/0000-0001-7814-5929"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhong Meng","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101709477","display_name":"Yu Wu","orcid":"https://orcid.org/0000-0002-5715-3011"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yu Wu","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016279564","display_name":"Naoyuki Kanda","orcid":"https://orcid.org/0000-0002-8628-3288"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Naoyuki Kanda","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101607148","display_name":"Liang Lu","orcid":"https://orcid.org/0000-0003-4005-679X"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Liang Lu","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101293966","display_name":"Xie Chen","orcid":"https://orcid.org/0000-0001-7423-617X"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xie Chen","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008578397","display_name":"Guoli Ye","orcid":null},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Guoli Ye","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090360853","display_name":"Eric Sun","orcid":"https://orcid.org/0000-0001-8805-9864"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Eric Sun","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100365053","display_name":"Jinyu Li","orcid":"https://orcid.org/0000-0002-1089-9748"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jinyu Li","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101928537","display_name":"Yifan Gong","orcid":"https://orcid.org/0000-0002-3912-097X"},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yifan Gong","raw_affiliation_strings":["Microsoft Corporation, Redmond, WA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Corporation, Redmond, WA, USA","institution_ids":["https://openalex.org/I1290206253"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":9,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.8191,"has_fulltext":false,"cited_by_count":14,"citation_normalized_percentile":{"value":0.87835648,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"2596","last_page":"2600"},"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":0.9968000054359436,"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/T10863","display_name":"Voice and Speech Disorders","score":0.9940999746322632,"subfield":{"id":"https://openalex.org/subfields/2737","display_name":"Physiology"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7097601294517517},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.641586422920227},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5114372968673706},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5110632181167603},{"id":"https://openalex.org/keywords/interpolation","display_name":"Interpolation (computer graphics)","score":0.49646979570388794},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.49103403091430664},{"id":"https://openalex.org/keywords/fusion","display_name":"Fusion","score":0.4709022343158722},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.46640291810035706},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.4318511486053467},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.42274224758148193},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.26440751552581787}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7097601294517517},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.641586422920227},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5114372968673706},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5110632181167603},{"id":"https://openalex.org/C137800194","wikidata":"https://www.wikidata.org/wiki/Q11713455","display_name":"Interpolation (computer graphics)","level":3,"score":0.49646979570388794},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.49103403091430664},{"id":"https://openalex.org/C158525013","wikidata":"https://www.wikidata.org/wiki/Q2593739","display_name":"Fusion","level":2,"score":0.4709022343158722},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.46640291810035706},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.4318511486053467},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42274224758148193},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.26440751552581787},{"id":"https://openalex.org/C104114177","wikidata":"https://www.wikidata.org/wiki/Q79782","display_name":"Motion (physics)","level":2,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/interspeech.2021-2075","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2021-2075","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Interspeech 2021","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.5099999904632568}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":55,"referenced_works":["https://openalex.org/W854541894","https://openalex.org/W1494198834","https://openalex.org/W1513820424","https://openalex.org/W1522301498","https://openalex.org/W1828163288","https://openalex.org/W1915251500","https://openalex.org/W1922655562","https://openalex.org/W1989549063","https://openalex.org/W2010362084","https://openalex.org/W2080005694","https://openalex.org/W2095705004","https://openalex.org/W2102113734","https://openalex.org/W2160815625","https://openalex.org/W2293634267","https://openalex.org/W2327501763","https://openalex.org/W2510867321","https://openalex.org/W2515801922","https://openalex.org/W2577366047","https://openalex.org/W2584667682","https://openalex.org/W2636483419","https://openalex.org/W2653985929","https://openalex.org/W2769025471","https://openalex.org/W2795867901","https://openalex.org/W2796339975","https://openalex.org/W2891980359","https://openalex.org/W2911629330","https://openalex.org/W2936252403","https://openalex.org/W2939164678","https://openalex.org/W2962760690","https://openalex.org/W2962784628","https://openalex.org/W2962824709","https://openalex.org/W2962894366","https://openalex.org/W2963747784","https://openalex.org/W2963970535","https://openalex.org/W2972621414","https://openalex.org/W2972818416","https://openalex.org/W3007227084","https://openalex.org/W3008008574","https://openalex.org/W3008037978","https://openalex.org/W3015194534","https://openalex.org/W3015915933","https://openalex.org/W3016234571","https://openalex.org/W3030437843","https://openalex.org/W3094667432","https://openalex.org/W3094979069","https://openalex.org/W3095311338","https://openalex.org/W3095697114","https://openalex.org/W3096815019","https://openalex.org/W3112702554","https://openalex.org/W3112742522","https://openalex.org/W3152221657","https://openalex.org/W3161873870","https://openalex.org/W3163560333","https://openalex.org/W4287634420","https://openalex.org/W4385245566"],"related_works":["https://openalex.org/W2916997151","https://openalex.org/W2949174760","https://openalex.org/W2594897229","https://openalex.org/W2151348424","https://openalex.org/W4221142855","https://openalex.org/W2050138804","https://openalex.org/W4290708361","https://openalex.org/W2129812225","https://openalex.org/W26527944","https://openalex.org/W2523799048"],"abstract_inverted_index":{"Integrating":[0],"external":[1,49],"language":[2,18],"models":[3,8],"(LMs)":[4],"into":[5],"end-to-end":[6],"(E2E)":[7],"remains":[9],"a":[10,37,66,113],"challenging":[11],"task":[12],"for":[13,99],"domain-adaptive":[14],"speech":[15],"recognition.Recently,":[16],"internal":[17,39,137],"model":[19,47,94],"estimation":[20],"(ILME)-based":[21],"LM":[22,40,50,61,83,100,138,146,154],"fusion":[23,84,123],"has":[24],"shown":[25],"significant":[26],"word":[27],"error":[28],"rate":[29],"(WER)":[30],"reduction":[31],"from":[32,42,159,177],"Shallow":[33,108],"Fusion":[34,109],"by":[35],"subtracting":[36],"weighted":[38],"score":[41],"an":[43,92],"interpolation":[44,62],"of":[45,91],"E2E":[46,93],"and":[48,69,130,172,179],"scores":[51],"during":[52,103],"beam":[53],"search.However,":[54],"on":[55,75,156,169,183],"different":[56,160,185],"test":[57,157,186],"sets,":[58],"the":[59,86,97,121],"optimal":[60],"weights":[63,101,147],"vary":[64],"over":[65],"wide":[67],"range":[68],"have":[70],"to":[71,95,126],"be":[72],"tuned":[73],"extensively":[74],"well-matched":[76],"validation":[77],"sets.In":[78],"this":[79],"work,":[80],"we":[81,111],"perform":[82],"in":[85,141,149],"minimum":[87],"WER":[88,175],"(MWER)":[89],"training":[90,106,116,151],"obviate":[96],"need":[98],"tuning":[102],"inference.Besides":[104],"MWER":[105,115,150,178],"with":[107,117,162],"(MWER-SF),":[110],"propose":[112],"novel":[114],"ILME":[118],"(MWER-ILME)":[119],"where":[120],"ILME-based":[122],"is":[124,134,139],"conducted":[125],"generate":[127],"N-best":[128],"hypotheses":[129],"their":[131],"posteriors.Additional":[132],"gradient":[133],"induced":[135],"when":[136],"engaged":[140],"MWER-ILME":[142,167],"loss":[143],"computation.During":[144],"inference,":[145],"pre-determined":[148],"enable":[152],"robust":[153],"integrations":[155],"sets":[158],"domains.Experimented":[161],"30K-hour":[163],"trained":[164],"transformer":[165],"transducers,":[166],"achieves":[168],"average":[170],"8.8%":[171],"5.8%":[173],"relative":[174],"reductions":[176],"MWER-SF":[180],"training,":[181],"respectively,":[182],"6":[184],"sets.":[187]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":7}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}