{"id":"https://openalex.org/W3162639446","doi":"https://doi.org/10.1109/icassp39728.2021.9414784","title":"Improved Neural Language Model Fusion for Streaming Recurrent Neural Network Transducer","display_name":"Improved Neural Language Model Fusion for Streaming Recurrent Neural Network Transducer","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3162639446","doi":"https://doi.org/10.1109/icassp39728.2021.9414784","mag":"3162639446"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9414784","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414784","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/A5062940343","display_name":"Suyoun Kim","orcid":"https://orcid.org/0000-0002-6822-337X"},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Suyoun Kim","raw_affiliation_strings":["Facebook AI,USA","Facebook AI, USA"],"affiliations":[{"raw_affiliation_string":"Facebook AI,USA","institution_ids":["https://openalex.org/I4210114444"]},{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047358828","display_name":"Yuan Shangguan","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yuan Shangguan","raw_affiliation_strings":["Facebook AI,USA","Facebook AI, USA"],"affiliations":[{"raw_affiliation_string":"Facebook AI,USA","institution_ids":["https://openalex.org/I4210114444"]},{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074237839","display_name":"Jay Mahadeokar","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jay Mahadeokar","raw_affiliation_strings":["Facebook AI,USA","Facebook AI, USA"],"affiliations":[{"raw_affiliation_string":"Facebook AI,USA","institution_ids":["https://openalex.org/I4210114444"]},{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091201956","display_name":"Antoine Bruguier","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Antoine Bruguier","raw_affiliation_strings":["Facebook AI,USA","Facebook AI, USA"],"affiliations":[{"raw_affiliation_string":"Facebook AI,USA","institution_ids":["https://openalex.org/I4210114444"]},{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047073253","display_name":"Christian Fuegen","orcid":null},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Christian Fuegen","raw_affiliation_strings":["Facebook AI,USA","Facebook AI, USA"],"affiliations":[{"raw_affiliation_string":"Facebook AI,USA","institution_ids":["https://openalex.org/I4210114444"]},{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041313589","display_name":"Michael L. Seltzer","orcid":"https://orcid.org/0000-0003-3474-2451"},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Michael L. Seltzer","raw_affiliation_strings":["Facebook AI,USA","Facebook AI, USA"],"affiliations":[{"raw_affiliation_string":"Facebook AI,USA","institution_ids":["https://openalex.org/I4210114444"]},{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5103098852","display_name":"Duc Le","orcid":"https://orcid.org/0000-0001-9490-2563"},"institutions":[{"id":"https://openalex.org/I4210114444","display_name":"Meta (United States)","ror":"https://ror.org/01zbnvs85","country_code":"US","type":"company","lineage":["https://openalex.org/I4210114444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Duc Le","raw_affiliation_strings":["Facebook AI,USA","Facebook AI, USA"],"affiliations":[{"raw_affiliation_string":"Facebook AI,USA","institution_ids":["https://openalex.org/I4210114444"]},{"raw_affiliation_string":"Facebook AI, USA","institution_ids":["https://openalex.org/I4210114444"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5062940343"],"corresponding_institution_ids":["https://openalex.org/I4210114444"],"apc_list":null,"apc_paid":null,"fwci":2.8553,"has_fulltext":false,"cited_by_count":24,"citation_normalized_percentile":{"value":0.92138989,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"7333","last_page":"7337"},"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9993000030517578,"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/T10028","display_name":"Topic Modeling","score":0.9973000288009644,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8465794324874878},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.8074456453323364},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.6723320484161377},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.6144267320632935},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5583544969558716},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.5561401844024658},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5194448828697205},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5011610984802246},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4966025948524475},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.4956752061843872},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4612864553928375},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.42668479681015015},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3672124445438385}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8465794324874878},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.8074456453323364},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.6723320484161377},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6144267320632935},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5583544969558716},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.5561401844024658},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5194448828697205},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5011610984802246},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4966025948524475},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.4956752061843872},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4612864553928375},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.42668479681015015},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3672124445438385},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","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/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9414784","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414784","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":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":37,"referenced_works":["https://openalex.org/W854541894","https://openalex.org/W1494198834","https://openalex.org/W1522301498","https://openalex.org/W1828163288","https://openalex.org/W1915251500","https://openalex.org/W2327501763","https://openalex.org/W2407080277","https://openalex.org/W2526425061","https://openalex.org/W2750499125","https://openalex.org/W2888779557","https://openalex.org/W2936774411","https://openalex.org/W2939111082","https://openalex.org/W2962704885","https://openalex.org/W2962760690","https://openalex.org/W2962824709","https://openalex.org/W2962826786","https://openalex.org/W2963240019","https://openalex.org/W2963250244","https://openalex.org/W2963362078","https://openalex.org/W2963827914","https://openalex.org/W2963979492","https://openalex.org/W2964084166","https://openalex.org/W2964121744","https://openalex.org/W3007227084","https://openalex.org/W3008037978","https://openalex.org/W3015686596","https://openalex.org/W3016010032","https://openalex.org/W3016234571","https://openalex.org/W3021984791","https://openalex.org/W3149629662","https://openalex.org/W6623517193","https://openalex.org/W6631190155","https://openalex.org/W6638749077","https://openalex.org/W6640059789","https://openalex.org/W6713762819","https://openalex.org/W6743477263","https://openalex.org/W6776544788"],"related_works":["https://openalex.org/W2160451571","https://openalex.org/W2495256954","https://openalex.org/W1566315437","https://openalex.org/W2259317772","https://openalex.org/W2594897229","https://openalex.org/W4221142855","https://openalex.org/W2151348424","https://openalex.org/W2050138804","https://openalex.org/W767271433","https://openalex.org/W4290708361"],"abstract_inverted_index":{"Recurrent":[0],"Neural":[1],"Network":[2],"Transducer":[3],"(RNN-T),":[4],"like":[5],"most":[6],"end-to-end":[7,41],"speech":[8],"recognition":[9],"model":[10,18],"architectures,":[11],"has":[12,31],"an":[13],"implicit":[14],"neural":[15],"network":[16],"language":[17],"(NNLM)":[19],"and":[20,66,92,140],"cannot":[21],"easily":[22],"leverage":[23],"unpaired":[24,127],"text":[25,128],"data":[26,129],"during":[27,63],"training.":[28],"Previous":[29],"work":[30,120],"proposed":[32],"various":[33],"fusion":[34,117],"methods":[35,85],"to":[36,43,53,59,80,108,130],"incorporate":[37],"external":[38,61],"NNLMs":[39,62,100],"into":[40],"ASR":[42],"address":[44],"this":[45,48],"weakness.":[46],"In":[47],"paper,":[49],"we":[50],"propose":[51],"extensions":[52],"these":[54],"techniques":[55],"that":[56],"allow":[57,93],"RNN-T":[58,133],"exploit":[60],"both":[64],"training":[65],"inference":[67],"time,":[68],"resulting":[69],"in":[70],"13-18%":[71],"relative":[72],"Word":[73],"Error":[74],"Rate":[75],"improvement":[76],"on":[77],"Librispeech":[78],"compared":[79],"strong":[81],"baselines.":[82],"Furthermore,":[83],"our":[84],"do":[86],"not":[87],"incur":[88],"extra":[89],"algorithmic":[90],"latency":[91],"for":[94,125],"flexible":[95],"plug-":[96],"and-play":[97],"of":[98,113],"different":[99,115],"without":[101],"re-training.":[102],"We":[103],"also":[104],"share":[105],"in-depth":[106],"analysis":[107],"better":[109],"understand":[110],"the":[111,114,136],"benefits":[112],"NNLM":[116],"methods.":[118],"Our":[119],"provides":[121],"a":[122],"reliable":[123],"technique":[124],"leveraging":[126],"significantly":[131],"improve":[132],"while":[134],"keeping":[135],"system":[137],"streamable,":[138],"flexible,":[139],"lightweight.":[141]},"counts_by_year":[{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":6},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":10}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}