{"id":"https://openalex.org/W3007037695","doi":"https://doi.org/10.1109/asru46091.2019.9003790","title":"A Density Ratio Approach to Language Model Fusion in End-to-End Automatic Speech Recognition","display_name":"A Density Ratio Approach to Language Model Fusion in End-to-End Automatic Speech Recognition","publication_year":2019,"publication_date":"2019-12-01","ids":{"openalex":"https://openalex.org/W3007037695","doi":"https://doi.org/10.1109/asru46091.2019.9003790","mag":"3007037695"},"language":"en","primary_location":{"id":"doi:10.1109/asru46091.2019.9003790","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru46091.2019.9003790","pdf_url":null,"source":{"id":"https://openalex.org/S4306498489","display_name":"2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2002.11268","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5030109648","display_name":"Erik McDermott","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":"Erik McDermott","raw_affiliation_strings":["Google Inc., USA","Google,,,,,"],"affiliations":[{"raw_affiliation_string":"Google Inc., USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google,,,,,","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108537573","display_name":"Ha\u015fim Sak","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":"Hasim Sak","raw_affiliation_strings":["Google Inc., USA","Google,,,,,"],"affiliations":[{"raw_affiliation_string":"Google Inc., USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google,,,,,","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5040131037","display_name":"Ehsan Variani","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":"Ehsan Variani","raw_affiliation_strings":["Google Inc., USA","Google,,,,,"],"affiliations":[{"raw_affiliation_string":"Google Inc., USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google,,,,,","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5030109648"],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":0.1447,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.61480189,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"434","last_page":"441"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":1.0,"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":1.0,"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":0.9983000159263611,"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":0.9983000159263611,"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":0.8012516498565674},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.7433305382728577},{"id":"https://openalex.org/keywords/hidden-markov-model","display_name":"Hidden Markov model","score":0.7272789478302002},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.7029327154159546},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6588653922080994},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.6148250102996826},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.5876177549362183},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5322867035865784},{"id":"https://openalex.org/keywords/sensor-fusion","display_name":"Sensor fusion","score":0.41607263684272766},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.37957534193992615},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.34812405705451965},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.06744614243507385}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8012516498565674},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.7433305382728577},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.7272789478302002},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.7029327154159546},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6588653922080994},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.6148250102996826},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.5876177549362183},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5322867035865784},{"id":"https://openalex.org/C33954974","wikidata":"https://www.wikidata.org/wiki/Q486494","display_name":"Sensor fusion","level":2,"score":0.41607263684272766},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.37957534193992615},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.34812405705451965},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.06744614243507385},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1109/asru46091.2019.9003790","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru46091.2019.9003790","pdf_url":null,"source":{"id":"https://openalex.org/S4306498489","display_name":"2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2002.11268","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2002.11268","pdf_url":"https://arxiv.org/pdf/2002.11268","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"},{"id":"mag:3007037695","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2002.11268.pdf","pdf_url":null,"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":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.2002.11268","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2002.11268","pdf_url":null,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2002.11268","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2002.11268","pdf_url":"https://arxiv.org/pdf/2002.11268","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":[{"score":0.5,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3007037695.pdf","grobid_xml":"https://content.openalex.org/works/W3007037695.grobid-xml"},"referenced_works_count":43,"referenced_works":["https://openalex.org/W65912994","https://openalex.org/W66978610","https://openalex.org/W179875071","https://openalex.org/W309335912","https://openalex.org/W811578723","https://openalex.org/W1553004968","https://openalex.org/W1560013842","https://openalex.org/W1828163288","https://openalex.org/W1915251500","https://openalex.org/W1990005915","https://openalex.org/W2033256038","https://openalex.org/W2064675550","https://openalex.org/W2112796928","https://openalex.org/W2114016253","https://openalex.org/W2127141656","https://openalex.org/W2160306971","https://openalex.org/W2293009711","https://openalex.org/W2327501763","https://openalex.org/W2577366047","https://openalex.org/W2592356860","https://openalex.org/W2748584437","https://openalex.org/W2787663903","https://openalex.org/W2886180730","https://openalex.org/W2888779557","https://openalex.org/W2939111082","https://openalex.org/W2951327905","https://openalex.org/W2962824709","https://openalex.org/W2963070863","https://openalex.org/W2963240019","https://openalex.org/W2963362078","https://openalex.org/W2964121744","https://openalex.org/W2964227577","https://openalex.org/W6602639124","https://openalex.org/W6602736415","https://openalex.org/W6610777176","https://openalex.org/W6631190155","https://openalex.org/W6638749077","https://openalex.org/W6640059789","https://openalex.org/W6696982659","https://openalex.org/W6728910023","https://openalex.org/W6743077814","https://openalex.org/W6752739165","https://openalex.org/W6753742271"],"related_works":["https://openalex.org/W3008037978","https://openalex.org/W3015760975","https://openalex.org/W2886180730","https://openalex.org/W3048653539","https://openalex.org/W2963311389","https://openalex.org/W3198800617","https://openalex.org/W2592356860","https://openalex.org/W2800057634","https://openalex.org/W1487889217","https://openalex.org/W2292359288","https://openalex.org/W281892562","https://openalex.org/W3090123943","https://openalex.org/W2612800309","https://openalex.org/W2486205537","https://openalex.org/W2395043307","https://openalex.org/W2891628540","https://openalex.org/W2566781703","https://openalex.org/W2767064007","https://openalex.org/W2736447239","https://openalex.org/W2157844668"],"abstract_inverted_index":{"This":[0],"article":[1],"describes":[2],"a":[3,23,33,36,41,60,102],"density":[4],"ratio":[5],"approach":[6,92,166],"to":[7,22,51,64,127,149,151,161,167],"integrating":[8],"external":[9],"Language":[10],"Models":[11],"(LMs)":[12],"into":[13],"end-to-end":[14,170],"models":[15],"for":[16,55,69,100,112,146],"Automatic":[17],"Speech":[18],"Recognition":[19],"(ASR).":[20],"Applied":[21],"Recurrent":[24],"Neural":[25,74],"Network":[26],"Transducer":[27],"(RNN-T)":[28],"ASR":[29,70,171],"model":[30,68,134],"trained":[31,135],"on":[32,72,136],"given":[34],"domain,":[35,58],"matched":[37],"in-domain":[38],"RNN-LM,":[39,44],"and":[40,97,169],"target":[42,57,106],"domain":[43,107],"the":[45,56,65,80,129,164],"proposed":[46,91],"method":[47,158],"uses":[48],"Bayes'":[49],"Rule":[50],"define":[52],"RNN-T":[53,133],"posteriors":[54],"in":[59,79,95],"manner":[61],"directly":[62],"analogous":[63],"classic":[66],"hybrid":[67],"based":[71],"Deep":[73],"Networks":[75],"(DNNs)":[76],"or":[77],"LSTMs":[78],"Hidden":[81],"Markov":[82],"Model":[83],"(HMM)":[84],"framework":[85],"(Bourlard":[86],"&":[87,139],"Morgan,":[88],"1994).":[89],"The":[90,155],"is":[93,110,144],"evaluated":[94,145],"cross-domain":[96],"limited-data":[98],"scenarios,":[99],"which":[101],"significant":[103],"amount":[104],"of":[105],"text":[108],"data":[109,123,141],"used":[111,126],"LM":[113,168],"training,":[114],"but":[115],"only":[116],"limited":[117],"(or":[118],"no)":[119],"{audio,":[120],"transcript}":[121],"training":[122],"pairs":[124],"are":[125],"train":[128],"RNN-T.":[130],"Specifically,":[131],"an":[132],"paired":[137],"audio":[138],"transcript":[140],"from":[142],"YouTube":[143],"its":[147],"ability":[148],"generalize":[150],"Voice":[152],"Search":[153],"data.":[154],"Density":[156],"Ratio":[157],"was":[159],"found":[160],"consistently":[162],"outperform":[163],"dominant":[165],"integration,":[172],"Shallow":[173],"Fusion.":[174]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2022,"cited_by_count":1}],"updated_date":"2026-03-21T08:13:44.787528","created_date":"2022-07-26T00:00:00"}