{"id":"https://openalex.org/W4372260124","doi":"https://doi.org/10.1109/icassp49357.2023.10095857","title":"On-the-Fly Text Retrieval for end-to-end ASR Adaptation","display_name":"On-the-Fly Text Retrieval for end-to-end ASR Adaptation","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4372260124","doi":"https://doi.org/10.1109/icassp49357.2023.10095857"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10095857","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10095857","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":"conference-paper","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/A5007614006","display_name":"Bolaji Yusuf","orcid":"https://orcid.org/0000-0001-9852-3456"},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]},{"id":"https://openalex.org/I4405392","display_name":"Bo\u011fazi\u00e7i University","ror":"https://ror.org/03z9tma90","country_code":"TR","type":"education","lineage":["https://openalex.org/I4405392"]},{"id":"https://openalex.org/I60587646","display_name":"Brno University of Technology","ror":"https://ror.org/03613d656","country_code":"CZ","type":"education","lineage":["https://openalex.org/I60587646"]}],"countries":["CZ","TR","US"],"is_corresponding":false,"raw_author_name":"Bolaji Yusuf","raw_affiliation_strings":["Amazon Alexa,USA","Faculty of Information Technology, Speech@FIT, Brno University of Technology, Czechia","Amazon Alexa, USA","Department of Electrical and Electronics Engineering, Bogazici University, Istanbul, Turkey"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon Alexa,USA","institution_ids":["https://openalex.org/I1311688040"]},{"raw_affiliation_string":"Faculty of Information Technology, Speech@FIT, Brno University of Technology, Czechia","institution_ids":["https://openalex.org/I60587646"]},{"raw_affiliation_string":"Amazon Alexa, USA","institution_ids":["https://openalex.org/I1311688040"]},{"raw_affiliation_string":"Department of Electrical and Electronics Engineering, Bogazici University, Istanbul, Turkey","institution_ids":["https://openalex.org/I4405392"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050696303","display_name":"Aditya Gourav","orcid":null},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Aditya Gourav","raw_affiliation_strings":["Amazon Alexa,USA","Amazon Alexa, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon Alexa,USA","institution_ids":["https://openalex.org/I1311688040"]},{"raw_affiliation_string":"Amazon Alexa, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014054572","display_name":"Ankur Gandhe","orcid":null},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ankur Gandhe","raw_affiliation_strings":["Amazon Alexa,USA","Amazon Alexa, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon Alexa,USA","institution_ids":["https://openalex.org/I1311688040"]},{"raw_affiliation_string":"Amazon Alexa, USA","institution_ids":["https://openalex.org/I1311688040"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5109934165","display_name":"Ivan Bulyko","orcid":null},"institutions":[{"id":"https://openalex.org/I1311688040","display_name":"Amazon (United States)","ror":"https://ror.org/04mv4n011","country_code":"US","type":"company","lineage":["https://openalex.org/I1311688040"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ivan Bulyko","raw_affiliation_strings":["Amazon Alexa,USA","Amazon Alexa, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Amazon Alexa,USA","institution_ids":["https://openalex.org/I1311688040"]},{"raw_affiliation_string":"Amazon Alexa, USA","institution_ids":["https://openalex.org/I1311688040"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"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":0.9998000264167786,"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":0.9998000264167786,"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":0.9997000098228455,"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":0.9997000098228455,"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":0.8260753154754639},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.741698145866394},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.5890865325927734},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5586061477661133},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5191832780838013},{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.49764207005500793},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.47259292006492615},{"id":"https://openalex.org/keywords/retraining","display_name":"Retraining","score":0.4524378776550293},{"id":"https://openalex.org/keywords/adapter","display_name":"Adapter (computing)","score":0.4480017125606537},{"id":"https://openalex.org/keywords/recall","display_name":"Recall","score":0.4450320303440094},{"id":"https://openalex.org/keywords/text-corpus","display_name":"Text corpus","score":0.4275706112384796}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8260753154754639},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.741698145866394},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.5890865325927734},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5586061477661133},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5191832780838013},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.49764207005500793},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.47259292006492615},{"id":"https://openalex.org/C2778712577","wikidata":"https://www.wikidata.org/wiki/Q3505966","display_name":"Retraining","level":2,"score":0.4524378776550293},{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.4480017125606537},{"id":"https://openalex.org/C100660578","wikidata":"https://www.wikidata.org/wiki/Q18733","display_name":"Recall","level":2,"score":0.4450320303440094},{"id":"https://openalex.org/C2474386","wikidata":"https://www.wikidata.org/wiki/Q461183","display_name":"Text corpus","level":2,"score":0.4275706112384796},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"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/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.0},{"id":"https://openalex.org/C155202549","wikidata":"https://www.wikidata.org/wiki/Q178803","display_name":"International trade","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49357.2023.10095857","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10095857","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"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.5400000214576721,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1828163288","https://openalex.org/W1915251500","https://openalex.org/W2124509324","https://openalex.org/W2577366047","https://openalex.org/W2886319145","https://openalex.org/W2949615363","https://openalex.org/W2963323070","https://openalex.org/W2963979492","https://openalex.org/W2972625221","https://openalex.org/W2998702515","https://openalex.org/W3008037978","https://openalex.org/W3016234571","https://openalex.org/W3034671305","https://openalex.org/W3097794466","https://openalex.org/W3163462603","https://openalex.org/W3163560333","https://openalex.org/W3196895794","https://openalex.org/W3197681820","https://openalex.org/W3198836239","https://openalex.org/W4205694376","https://openalex.org/W4224918838","https://openalex.org/W4225308107","https://openalex.org/W4225985539","https://openalex.org/W4226093136","https://openalex.org/W4288087322","https://openalex.org/W4298422451","https://openalex.org/W4319862261","https://openalex.org/W6632455782","https://openalex.org/W6638749077","https://openalex.org/W6640059789","https://openalex.org/W6727099177","https://openalex.org/W6770813465","https://openalex.org/W6779857854","https://openalex.org/W6849799735"],"related_works":["https://openalex.org/W2050078012","https://openalex.org/W2133028525","https://openalex.org/W4306381730","https://openalex.org/W4229060448","https://openalex.org/W2006651773","https://openalex.org/W2027050655","https://openalex.org/W3028244590","https://openalex.org/W2981692913","https://openalex.org/W2014369232","https://openalex.org/W3122042562"],"abstract_inverted_index":{"End-to-end":[0],"speech":[1],"recognition":[2,137],"models":[3,21],"are":[4,81,148],"improved":[5],"by":[6,12,87,141],"incorporating":[7],"external":[8,16,69],"text":[9,70],"sources,":[10],"typically":[11],"fusion":[13,135],"with":[14,59],"an":[15,68,88],"language":[17,20,62],"model.":[18],"Such":[19],"have":[22],"to":[23,47,154],"be":[24,45,101],"retrained":[25],"whenever":[26],"the":[27,36,96,105,114,119,146,150],"corpus":[28,38,71,97],"of":[29,98,108,121,128,138],"interest":[30,99],"changes.":[31],"Furthermore,":[32],"since":[33],"they":[34],"store":[35],"entire":[37],"in":[39],"their":[40],"parameters,":[41],"rare":[42],"words":[43],"can":[44,100],"challenging":[46],"recall.":[48],"In":[49],"this":[50],"work,":[51],"we":[52],"propose":[53],"augmenting":[54],"a":[55,60,75,122,126],"transducer-based":[56],"ASR":[57,77],"model":[58,116],"retrieval":[61],"model,":[63],"which":[64,90],"directly":[65],"retrieves":[66],"from":[67],"plausible":[72],"completions":[73,80],"for":[74],"partial":[76],"hypothesis.":[78],"These":[79],"then":[82],"integrated":[83],"into":[84],"subsequent":[85],"predictions":[86],"adapter,":[89],"is":[91],"trained":[92],"once,":[93],"so":[94],"that":[95,113],"switched":[102],"without":[103],"incurring":[104],"computational":[106],"overhead":[107],"retraining.":[109],"Our":[110],"experiments":[111],"show":[112],"proposed":[115],"significantly":[117],"improves":[118],"performance":[120],"transducer":[123],"baseline":[124],"on":[125,136],"pair":[127],"question-answering":[129],"datasets.":[130],"Further,":[131],"it":[132],"outperforms":[133],"shallow":[134],"named":[139],"entities":[140],"about":[142],"7%":[143],"relative;":[144],"when":[145],"two":[147],"combined,":[149],"relative":[151],"improvement":[152],"increases":[153],"13%.":[155]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1}],"updated_date":"2026-07-27T08:26:11.824852","created_date":"2025-10-10T00:00:00"}