{"id":"https://openalex.org/W3160987270","doi":"https://doi.org/10.1109/icassp39728.2021.9414911","title":"Multitask Learning and Joint Optimization for Transformer-RNN-Transducer Speech Recognition","display_name":"Multitask Learning and Joint Optimization for Transformer-RNN-Transducer Speech Recognition","publication_year":2021,"publication_date":"2021-05-13","ids":{"openalex":"https://openalex.org/W3160987270","doi":"https://doi.org/10.1109/icassp39728.2021.9414911","mag":"3160987270"},"language":"en","primary_location":{"id":"doi:10.1109/icassp39728.2021.9414911","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414911","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/A5038521076","display_name":"Jae-Jin Jeon","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jae-Jin Jeon","raw_affiliation_strings":["AI R&D Team, Kakao Enterprise, Seongnam-si, Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI R&D Team, Kakao Enterprise, Seongnam-si, Korea","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5013285390","display_name":"Eesung Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Eesung Kim","raw_affiliation_strings":["AI R&D Team, Kakao Enterprise, Seongnam-si, Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI R&D Team, Kakao Enterprise, Seongnam-si, Korea","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.6792,"has_fulltext":false,"cited_by_count":14,"citation_normalized_percentile":{"value":0.86796222,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"6793","last_page":"6797"},"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":0.9990000128746033,"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/T11309","display_name":"Music and Audio Processing","score":0.996399998664856,"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":0.7944925427436829},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.7698752880096436},{"id":"https://openalex.org/keywords/recurrent-neural-network","display_name":"Recurrent neural network","score":0.683962881565094},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.5857439041137695},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5439605116844177},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5390830039978027},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.5053170323371887},{"id":"https://openalex.org/keywords/multi-task-learning","display_name":"Multi-task learning","score":0.4857233166694641},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4630778729915619},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4570049047470093},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.42904970049858093},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3497996926307678},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.12634649872779846},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.08902397751808167},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.08565202355384827},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.07199975848197937}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7944925427436829},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7698752880096436},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.683962881565094},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5857439041137695},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5439605116844177},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5390830039978027},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.5053170323371887},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.4857233166694641},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4630778729915619},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4570049047470093},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.42904970049858093},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3497996926307678},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.12634649872779846},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.08902397751808167},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.08565202355384827},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.07199975848197937},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp39728.2021.9414911","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp39728.2021.9414911","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":30,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1522301498","https://openalex.org/W1828163288","https://openalex.org/W1855892484","https://openalex.org/W2127141656","https://openalex.org/W2936774411","https://openalex.org/W2941814890","https://openalex.org/W2943845043","https://openalex.org/W2946200149","https://openalex.org/W2962780374","https://openalex.org/W2963250244","https://openalex.org/W2963403868","https://openalex.org/W2963414781","https://openalex.org/W2964121744","https://openalex.org/W2970730223","https://openalex.org/W2982413405","https://openalex.org/W3016010032","https://openalex.org/W3021469861","https://openalex.org/W3025165719","https://openalex.org/W3095173472","https://openalex.org/W3097777922","https://openalex.org/W4288088457","https://openalex.org/W4385245566","https://openalex.org/W6631190155","https://openalex.org/W6638749077","https://openalex.org/W6639156005","https://openalex.org/W6739901393","https://openalex.org/W6761563299","https://openalex.org/W6763832098","https://openalex.org/W6769806307"],"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":{"Recently,":[0],"several":[1],"types":[2],"of":[3,16,80],"end-to-end":[4],"speech":[5],"recognition":[6],"methods":[7,54,60,110],"named":[8],"transformer-transducer":[9],"were":[10],"introduced.":[11],"According":[12],"to":[13],"those":[14],"kinds":[15],"methods,":[17],"transcription":[18],"networks":[19,29,39],"are":[20],"generally":[21],"modeled":[22,32],"by":[23,33,90,117],"transformer-based":[24],"neural":[25,38],"networks,":[26],"while":[27],"prediction":[28],"could":[30],"be":[31],"either":[34],"transformers":[35],"or":[36],"recurrent":[37],"(RNN).":[40],"In":[41],"this":[42],"paper,":[43],"we":[44],"propose":[45],"novel":[46],"multitask":[47],"learning,":[48],"joint":[49,52],"optimization,":[50],"and":[51,120,125],"decoding":[53],"for":[55,98,123],"transformer-RNN-transducer":[56],"systems.":[57],"Our":[58],"proposed":[59,109],"have":[61],"the":[62,67,73,78,94,99,108,131],"main":[63],"advantage":[64],"in":[65],"that":[66,107],"model":[68,84,133],"can":[69,111],"maintain":[70],"information":[71],"on":[72],"large":[74],"text":[75],"corpus":[76],"eliminating":[77],"necessity":[79],"an":[81,137],"external":[82,138],"language":[83],"(LM).":[85],"We":[86,104],"prove":[87],"their":[88],"effectiveness":[89],"performing":[91],"experiments":[92],"utilizing":[93],"well-known":[95],"ESPNET":[96],"toolkit":[97],"widely":[100],"used":[101],"Librispeech":[102],"datasets.":[103],"also":[105],"show":[106],"reduce":[112],"word":[113],"error":[114],"rate":[115],"(WER)":[116],"16.6":[118],"%":[119,122],"13.3":[121],"test-clean":[124],"test-other":[126],"datasets,":[127],"respectively,":[128],"without":[129],"changing":[130],"overall":[132],"structure":[134],"nor":[135],"exploiting":[136],"LM.":[139]},"counts_by_year":[{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":3},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}