{"id":"https://openalex.org/W4225308107","doi":"https://doi.org/10.1109/icassp43922.2022.9747862","title":"Integrating Text Inputs for Training and Adapting RNN Transducer ASR Models","display_name":"Integrating Text Inputs for Training and Adapting RNN Transducer ASR Models","publication_year":2022,"publication_date":"2022-04-27","ids":{"openalex":"https://openalex.org/W4225308107","doi":"https://doi.org/10.1109/icassp43922.2022.9747862"},"language":"en","primary_location":{"id":"doi:10.1109/icassp43922.2022.9747862","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9747862","pdf_url":null,"source":{"id":"https://openalex.org/S4363607702","display_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2022 - 2022 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/A5101787514","display_name":"Samuel Thomas","orcid":"https://orcid.org/0000-0001-7573-0620"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Samuel Thomas","raw_affiliation_strings":["IBM Research AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003725957","display_name":"Brian Kingsbury","orcid":"https://orcid.org/0000-0002-1343-6837"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Brian Kingsbury","raw_affiliation_strings":["IBM Research AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079994647","display_name":"George Saon","orcid":"https://orcid.org/0009-0004-6837-5009"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"George Saon","raw_affiliation_strings":["IBM Research AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research AI","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110252428","display_name":"Hong-Kwang Jeff Kuo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hong-Kwang J. Kuo","raw_affiliation_strings":["IBM Research AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"IBM Research AI","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":23,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"8127","last_page":"8131"},"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/T12031","display_name":"Speech and dialogue systems","score":0.9955000281333923,"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":0.9952999949455261,"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":0.8267903327941895},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.7236446738243103},{"id":"https://openalex.org/keywords/nist","display_name":"NIST","score":0.6583213806152344},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.6370739936828613},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.6079505085945129},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.6000185608863831},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5208176374435425},{"id":"https://openalex.org/keywords/personalization","display_name":"Personalization","score":0.5062681436538696},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4889124929904938},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.46571826934814453},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.43805402517318726},{"id":"https://openalex.org/keywords/modular-design","display_name":"Modular design","score":0.4266047775745392},{"id":"https://openalex.org/keywords/hidden-markov-model","display_name":"Hidden Markov model","score":0.4212607145309448},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.4191702902317047},{"id":"https://openalex.org/keywords/acoustic-model","display_name":"Acoustic model","score":0.41614678502082825},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.33665919303894043},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.21386852860450745},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.15611833333969116}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8267903327941895},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.7236446738243103},{"id":"https://openalex.org/C111219384","wikidata":"https://www.wikidata.org/wiki/Q6954384","display_name":"NIST","level":2,"score":0.6583213806152344},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.6370739936828613},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.6079505085945129},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.6000185608863831},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5208176374435425},{"id":"https://openalex.org/C183003079","wikidata":"https://www.wikidata.org/wiki/Q1000371","display_name":"Personalization","level":2,"score":0.5062681436538696},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4889124929904938},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.46571826934814453},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.43805402517318726},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.4266047775745392},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.4212607145309448},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.4191702902317047},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.41614678502082825},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.33665919303894043},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.21386852860450745},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.15611833333969116},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","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/C94625758","wikidata":"https://www.wikidata.org/wiki/Q7163","display_name":"Politics","level":2,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"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/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"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/C97355855","wikidata":"https://www.wikidata.org/wiki/Q11473","display_name":"Thermodynamics","level":1,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp43922.2022.9747862","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9747862","pdf_url":null,"source":{"id":"https://openalex.org/S4363607702","display_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W1828163288","https://openalex.org/W2077302143","https://openalex.org/W2936774411","https://openalex.org/W2937780860","https://openalex.org/W2962760690","https://openalex.org/W2963414781","https://openalex.org/W2973127116","https://openalex.org/W3007227084","https://openalex.org/W3008037978","https://openalex.org/W3015686596","https://openalex.org/W3095384101","https://openalex.org/W3100460087","https://openalex.org/W3152221657","https://openalex.org/W3163300396","https://openalex.org/W3196869722","https://openalex.org/W3197314576","https://openalex.org/W3197661863","https://openalex.org/W6638749077","https://openalex.org/W6747398299","https://openalex.org/W6784579990"],"related_works":["https://openalex.org/W2149402184","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/W2244609359"],"abstract_inverted_index":{"Compared":[0],"to":[1,21,32,35,99,135],"hybrid":[2],"automatic":[3],"speech":[4,84,96],"recognition":[5],"(ASR)":[6],"systems":[7],"that":[8,61],"use":[9],"a":[10,22,45,62,90],"modular":[11],"architecture":[12],"in":[13,148],"which":[14],"each":[15],"component":[16,70],"can":[17,71],"be":[18,72],"in-dependently":[19],"adapted":[20,74],"new":[23,165],"domain,":[24],"recent":[25],"end-to-end":[26],"(E2E)":[27],"ASR":[28,54],"system":[29],"are":[30],"harder":[31],"customize":[33],"due":[34],"their":[36],"all-neural":[37],"monolithic":[38],"construction.":[39],"In":[40],"this":[41,57,130,152],"paper,":[42],"we":[43,59],"propose":[44],"novel":[46,153],"text":[47,86,161],"representation":[48],"and":[49,85,109],"training":[50],"framework":[51],"for":[52],"E2E":[53],"models.":[55],"With":[56],"approach,":[58],"show":[60],"trained":[63,81,93],"RNN":[64],"Transducer":[65],"(RNN-T)":[66],"model\u2019s":[67],"internal":[68],"LM":[69,154],"effectively":[73],"with":[75,97,151],"text-only":[76],"data.":[77],"An":[78],"RNN-T":[79,133],"model":[80,92,134],"using":[82,158],"both":[83],"inputs":[87],"improves":[88],"over":[89],"baseline":[91],"on":[94,106],"just":[95],"close":[98],"13%":[100],"word":[101,143],"error":[102,144],"rate":[103,145],"(WER)":[104,146],"reduction":[105,147],"the":[107,114,122,164],"Switchboard":[108],"CallHome":[110],"test":[111],"sets":[112],"of":[113,121],"NIST":[115],"Hub5":[116],"2000":[117],"evaluation.":[118],"The":[119],"usefulness":[120],"proposed":[123],"approach":[124],"is":[125],"further":[126],"demonstrated":[127],"by":[128],"customizing":[129],"general":[131],"purpose":[132],"three":[136],"separate":[137],"datasets.":[138],"We":[139],"observe":[140],"20-45%":[141],"relative":[142],"these":[149],"settings":[150],"style":[155],"customization":[156],"technique":[157],"only":[159],"unpaired":[160],"data":[162],"from":[163],"domains.":[166]},"counts_by_year":[{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":14}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}