{"id":"https://openalex.org/W4392902792","doi":"https://doi.org/10.1109/icassp48485.2024.10447399","title":"Efficient Adapter Finetuning for Tail Languages in Streaming Multilingual ASR","display_name":"Efficient Adapter Finetuning for Tail Languages in Streaming Multilingual ASR","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392902792","doi":"https://doi.org/10.1109/icassp48485.2024.10447399"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10447399","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447399","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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/A5002187986","display_name":"Junwen Bai","orcid":"https://orcid.org/0000-0001-7939-4927"},"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":"Junwen Bai","raw_affiliation_strings":["Google,USA","Google, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100677409","display_name":"Bo Li","orcid":"https://orcid.org/0000-0003-4883-7267"},"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":"Bo Li","raw_affiliation_strings":["Google,USA","Google, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030757951","display_name":"Qiujia Li","orcid":"https://orcid.org/0000-0003-3074-3692"},"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":"Qiujia Li","raw_affiliation_strings":["Google,USA","Google, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070513394","display_name":"Tara N. Sainath","orcid":"https://orcid.org/0000-0002-4126-6556"},"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":"Tara N. Sainath","raw_affiliation_strings":["Google,USA","Google, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5032931723","display_name":"Trevor Strohman","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":"Trevor Strohman","raw_affiliation_strings":["Google,USA","Google, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Google,USA","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"Google, USA","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":1.778,"has_fulltext":false,"cited_by_count":5,"citation_normalized_percentile":{"value":0.85880462,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"10841","last_page":"10845"},"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/T10863","display_name":"Voice and Speech Disorders","score":0.9961000084877014,"subfield":{"id":"https://openalex.org/subfields/2737","display_name":"Physiology"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9958999752998352,"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":0.8685630559921265},{"id":"https://openalex.org/keywords/asynchronous-communication","display_name":"Asynchronous communication","score":0.6931395530700684},{"id":"https://openalex.org/keywords/adapter","display_name":"Adapter (computing)","score":0.6897473931312561},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5478739738464355},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.41933026909828186},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3751320242881775},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.3615313768386841},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.15105542540550232},{"id":"https://openalex.org/keywords/computer-hardware","display_name":"Computer hardware","score":0.09572851657867432}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8685630559921265},{"id":"https://openalex.org/C151319957","wikidata":"https://www.wikidata.org/wiki/Q752739","display_name":"Asynchronous communication","level":2,"score":0.6931395530700684},{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.6897473931312561},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5478739738464355},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.41933026909828186},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3751320242881775},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3615313768386841},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.15105542540550232},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.09572851657867432}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10447399","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447399","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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","id":"https://metadata.un.org/sdg/4","score":0.5899999737739563}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W2144499799","https://openalex.org/W2744999500","https://openalex.org/W2928941594","https://openalex.org/W2936774411","https://openalex.org/W2971840980","https://openalex.org/W3008434450","https://openalex.org/W3016234571","https://openalex.org/W3026041220","https://openalex.org/W3035160371","https://openalex.org/W3036878841","https://openalex.org/W3096032230","https://openalex.org/W3097777922","https://openalex.org/W3160766462","https://openalex.org/W3161686170","https://openalex.org/W3163907627","https://openalex.org/W3165404421","https://openalex.org/W3173563729","https://openalex.org/W3197845195","https://openalex.org/W3198429080","https://openalex.org/W3212799896","https://openalex.org/W3213029956","https://openalex.org/W4210463634","https://openalex.org/W4221155340","https://openalex.org/W4225295099","https://openalex.org/W4225307083","https://openalex.org/W4296069150","https://openalex.org/W4319862468","https://openalex.org/W4323066695","https://openalex.org/W4375869211","https://openalex.org/W4378105483","https://openalex.org/W4385823078","https://openalex.org/W4385823224","https://openalex.org/W4391021732","https://openalex.org/W6631190155","https://openalex.org/W6759579507","https://openalex.org/W6760633627","https://openalex.org/W6780226713","https://openalex.org/W6810259195","https://openalex.org/W6850218400","https://openalex.org/W6852909395"],"related_works":["https://openalex.org/W2133028525","https://openalex.org/W4229060448","https://openalex.org/W4306381730","https://openalex.org/W2981692913","https://openalex.org/W3044188621","https://openalex.org/W3184035966","https://openalex.org/W2485605994","https://openalex.org/W2160602540","https://openalex.org/W4385571108","https://openalex.org/W4321593827"],"abstract_inverted_index":{"The":[0,123,157,170],"end-to-end":[1],"ASR":[2],"model":[3,82,132,142,159,171,224],"is":[4,14,136,144,173],"often":[5],"desired":[6],"in":[7,118],"the":[8,31,61,71,81,119,130,139,145,150,161,219,222,229],"streaming":[9,120],"multilingual":[10,121,178],"scenario":[11],"since":[12],"it":[13],"easier":[15],"to":[16,46,78,88,204],"deploy":[17],"and":[18,34,143,202],"can":[19,217],"benefit":[20],"from":[21,164],"pre-trained":[22],"speech":[23],"models":[24],"such":[25],"as":[26,67],"powerful":[27],"foundation":[28,141],"models.":[29],"Meanwhile,":[30],"heterogeneous":[32],"nature":[33],"imbalanced":[35],"data":[36,62],"abundance":[37],"of":[38,70,129,221],"different":[39,51,165,168],"languages":[40,52,117,185],"may":[41,64],"cause":[42],"performance":[43,49,172,232],"degradation,":[44],"leading":[45],"asynchronous":[47,230],"peak":[48,231],"for":[50,115,127,167],"during":[53,149],"training,":[54],"especially":[55],"on":[56,175,200,206],"tail":[57,116,184],"ones.":[58],"Sometimes":[59],"even":[60],"itself":[63],"become":[65],"unavailable":[66],"a":[68,106,176,207],"result":[69],"enhanced":[72,111],"privacy":[73],"protection.":[74],"Existing":[75],"work":[76],"tends":[77],"significantly":[79],"increase":[80],"size":[83],"or":[84],"learn":[85],"language-specific":[86],"decoders":[87],"accommodate":[89],"each":[90],"language":[91],"separately.":[92],"In":[93],"this":[94],"study,":[95],"we":[96,211],"explore":[97],"simple":[98],"yet":[99],"effective":[100],"Language-Dependent":[101],"Adapter":[102],"(LDA)":[103],"finetuning":[104,151],"under":[105],"cascaded":[107],"Conformer":[108],"transducer":[109],"framework":[110],"by":[112],"teacher":[113],"pseudolabeling":[114],"ASR.":[122],"adapter":[124,162],"only":[125,146],"accounts":[126],"0.4%":[128],"full":[131,223],"per":[133],"language.":[134],"It":[135],"plugged":[137],"into":[138],"frozen":[140],"trainable":[147],"module":[148],"process":[152],"with":[153],"noisy":[154],"student":[155],"training.":[156],"final":[158],"merges":[160],"parameters":[163],"checkpoints":[166],"languages.":[169],"validated":[174],"challenging":[177],"dictation":[179],"dataset,":[180],"which":[181],"includes":[182],"39":[183],"across":[186],"Latin,":[187],"Greek,":[188],"Arabic,":[189],"etc.":[190],"Our":[191],"proposed":[192],"method":[193],"brings":[194],"12.2%":[195],"word":[196],"error":[197],"rate":[198],"reduction":[199],"average":[201],"up":[203],"37.5%":[205],"single":[208],"locale.":[209],"Furthermore,":[210],"show":[212],"that":[213],"our":[214],"parameter-efficient":[215],"LDA":[216],"match":[218],"quality":[220],"finetuning,":[225],"thus":[226],"greatly":[227],"alleviating":[228],"issue.":[233]},"counts_by_year":[{"year":2025,"cited_by_count":5}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}