{"id":"https://openalex.org/W4375869211","doi":"https://doi.org/10.1109/icassp49357.2023.10096330","title":"Efficient Domain Adaptation for Speech Foundation Models","display_name":"Efficient Domain Adaptation for Speech Foundation Models","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4375869211","doi":"https://doi.org/10.1109/icassp49357.2023.10096330"},"language":"en","primary_location":{"is_oa":true,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10096330","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},"type":"article","type_crossref":"proceedings-article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.1109/icassp49357.2023.10096330","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100677404","display_name":"Bo Li","orcid":"https://orcid.org/0000-0003-1483-8077"},"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 LLC, USA"],"affiliations":[{"raw_affiliation_string":"Google LLC, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014088530","display_name":"Dongseong Hwang","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":"Dongseong Hwang","raw_affiliation_strings":["Google LLC, USA"],"affiliations":[{"raw_affiliation_string":"Google LLC, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069872293","display_name":"Zhouyuan Huo","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":"Zhouyuan Huo","raw_affiliation_strings":["Google LLC, USA"],"affiliations":[{"raw_affiliation_string":"Google LLC, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","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 LLC, USA"],"affiliations":[{"raw_affiliation_string":"Google LLC, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057360812","display_name":"Guru Prakash","orcid":"https://orcid.org/0000-0003-0884-709X"},"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":"Guru Prakash","raw_affiliation_strings":["Google LLC, USA"],"affiliations":[{"raw_affiliation_string":"Google LLC, 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 LLC, USA"],"affiliations":[{"raw_affiliation_string":"Google LLC, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032798707","display_name":"Khe Chai Sim","orcid":"https://orcid.org/0000-0002-0866-2223"},"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":"Khe Chai Sim","raw_affiliation_strings":["Google LLC, USA"],"affiliations":[{"raw_affiliation_string":"Google LLC, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092339709","display_name":"Yu Zhang","orcid":"https://orcid.org/0009-0007-7826-1306"},"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":"Yu Zhang","raw_affiliation_strings":["Google LLC, USA"],"affiliations":[{"raw_affiliation_string":"Google LLC, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100750907","display_name":"Wei Han","orcid":"https://orcid.org/0000-0002-4201-9645"},"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":"Wei Han","raw_affiliation_strings":["Google LLC, USA"],"affiliations":[{"raw_affiliation_string":"Google LLC, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","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 LLC, USA"],"affiliations":[{"raw_affiliation_string":"Google LLC, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110711540","display_name":"Fran\u00e7oise Beaufays","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":"Francoise Beaufays","raw_affiliation_strings":["Google LLC, USA"],"affiliations":[{"raw_affiliation_string":"Google LLC, USA","institution_ids":["https://openalex.org/I1291425158"]}]}],"institution_assertions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":4.534,"has_fulltext":false,"cited_by_count":11,"citation_normalized_percentile":{"value":0.999754,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9999,"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":0.9999,"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":0.998,"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"}},{"id":"https://openalex.org/T11309","display_name":"Music and Audio Processing","score":0.9961,"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/adapter","display_name":"Adapter (computing)","score":0.60667527},{"id":"https://openalex.org/keywords/domain-adaptation","display_name":"Domain Adaptation","score":0.4109712}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8882959},{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.60667527},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.5725888},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.51288867},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.48232934},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46086583},{"id":"https://openalex.org/C2776434776","wikidata":"https://www.wikidata.org/wiki/Q19246213","display_name":"Domain adaptation","level":3,"score":0.4109712},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"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/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"is_oa":true,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10096330","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},{"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2302.01496","pdf_url":"https://arxiv.org/pdf/2302.01496","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":["Cornell University"],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false}],"best_oa_location":{"is_oa":true,"landing_page_url":"https://doi.org/10.1109/icassp49357.2023.10096330","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},"sustainable_development_goals":[],"grants":[],"datasets":[],"versions":[],"referenced_works_count":44,"referenced_works":["https://openalex.org/W1828163288","https://openalex.org/W1855892484","https://openalex.org/W2033256038","https://openalex.org/W2127141656","https://openalex.org/W2747467128","https://openalex.org/W2889012072","https://openalex.org/W2896457183","https://openalex.org/W2928941594","https://openalex.org/W2931364255","https://openalex.org/W2936774411","https://openalex.org/W2963122170","https://openalex.org/W2963362078","https://openalex.org/W3008181812","https://openalex.org/W3008898571","https://openalex.org/W3026041220","https://openalex.org/W3035160371","https://openalex.org/W3097777922","https://openalex.org/W3099793224","https://openalex.org/W3100270690","https://openalex.org/W3139918052","https://openalex.org/W3149629662","https://openalex.org/W3195577433","https://openalex.org/W3198004110","https://openalex.org/W3198694222","https://openalex.org/W3202037040","https://openalex.org/W3204696009","https://openalex.org/W3204917342","https://openalex.org/W3205949070","https://openalex.org/W3207222250","https://openalex.org/W3207300132","https://openalex.org/W3212799896","https://openalex.org/W4210463634","https://openalex.org/W4221161761","https://openalex.org/W4224939570","https://openalex.org/W4225295099","https://openalex.org/W4225529283","https://openalex.org/W4226033575","https://openalex.org/W4226120743","https://openalex.org/W4287900772","https://openalex.org/W4288089799","https://openalex.org/W4296070216","https://openalex.org/W4297808394","https://openalex.org/W4307322847","https://openalex.org/W4385245566"],"related_works":["https://openalex.org/W4389474468","https://openalex.org/W4321649381","https://openalex.org/W4300172004","https://openalex.org/W4295929828","https://openalex.org/W3203792196","https://openalex.org/W3180787869","https://openalex.org/W3156096827","https://openalex.org/W2997645659","https://openalex.org/W2955455867","https://openalex.org/W2133028525"],"abstract_inverted_index":{"Foundation":[0],"models":[1,44],"(FMs),":[2],"that":[3],"are":[4,12,104],"trained":[5,159],"on":[6,162],"broad":[7],"data":[8,33,134,149],"at":[9],"scale":[10],"and":[11,40,49,80,93,102,123,135,150],"adaptable":[13],"to":[14,107,131,155],"a":[15,58,112,120],"wide":[16],"range":[17],"of":[18,115],"downstream":[19],"tasks,":[20],"have":[21,45],"brought":[22],"large":[23],"interest":[24],"in":[25],"the":[26,31,73,82,108,141,156],"research":[27],"community.":[28],"Benefiting":[29],"from":[30,160],"diverse":[32],"sources":[34],"such":[35],"as":[36],"different":[37],"modalities,":[38],"languages":[39],"application":[41],"domains,":[42],"foundation":[43],"demonstrated":[46],"strong":[47],"generalization":[48],"knowledge":[50],"transfer":[51],"capabilities.":[52],"In":[53],"this":[54],"paper,":[55],"we":[56],"present":[57],"pioneering":[59],"study":[60],"towards":[61],"building":[62],"an":[63],"efficient":[64],"solution":[65],"for":[66,78,88],"FM-based":[67],"speech":[68],"recognition":[69],"systems.":[70],"We":[71],"adopt":[72],"recently":[74],"developed":[75],"self-supervised":[76],"BEST-RQ":[77],"pretraining,":[79],"extend":[81],"joint":[83],"training":[84],"strategy":[85],"JUST":[86],"Hydra":[87],"finetuning":[89],"using":[90],"both":[91,133],"source":[92],"unsuper-vised":[94],"target":[95,109],"domain":[96,110],"data.":[97,118,167],"The":[98],"FM":[99],"encoder":[100],"adapter":[101],"decoder":[103],"then":[105],"finetuned":[106,152],"with":[111,144],"small":[113],"amount":[114],"super-vised":[116],"in-domain":[117,148,166],"On":[119],"large-scale":[121],"YouTube":[122],"Voice":[124],"Search":[125],"task,":[126],"our":[127],"method":[128],"is":[129],"shown":[130],"be":[132],"model":[136,158],"parameter":[137],"efficient.":[138],"It":[139],"achieves":[140],"same":[142],"quality":[143],"only":[145],"21.6M":[146],"supervised":[147,165],"130.8M":[151],"parameters,":[153],"compared":[154],"731.1M":[157],"scratch":[161],"additional":[163],"300M":[164]},"cited_by_api_url":"https://api.openalex.org/works?filter=cites:W4375869211","counts_by_year":[{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":6}],"updated_date":"2024-12-13T04:10:01.172493","created_date":"2023-05-10"}