{"id":"https://openalex.org/W4289824098","doi":"https://doi.org/10.1109/jstsp.2022.3195367","title":"Momentum Pseudo-Labeling: Semi-Supervised ASR With Continuously Improving Pseudo-Labels","display_name":"Momentum Pseudo-Labeling: Semi-Supervised ASR With Continuously Improving Pseudo-Labels","publication_year":2022,"publication_date":"2022-08-05","ids":{"openalex":"https://openalex.org/W4289824098","doi":"https://doi.org/10.1109/jstsp.2022.3195367"},"language":"en","primary_location":{"id":"doi:10.1109/jstsp.2022.3195367","is_oa":true,"landing_page_url":"https://doi.org/10.1109/jstsp.2022.3195367","pdf_url":"https://ieeexplore.ieee.org/ielx7/4200690/9923627/09846970.pdf","source":{"id":"https://openalex.org/S42167783","display_name":"IEEE Journal of Selected Topics in Signal Processing","issn_l":"1932-4553","issn":["1932-4553","1941-0484"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Journal of Selected Topics in Signal Processing","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://ieeexplore.ieee.org/ielx7/4200690/9923627/09846970.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102017467","display_name":"Yosuke Higuchi","orcid":"https://orcid.org/0000-0003-4500-8957"},"institutions":[{"id":"https://openalex.org/I150744194","display_name":"Waseda University","ror":"https://ror.org/00ntfnx83","country_code":"JP","type":"education","lineage":["https://openalex.org/I150744194"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Yosuke Higuchi","raw_affiliation_strings":["Waseda University, Tokyo, Japan"],"raw_orcid":"https://orcid.org/0000-0003-4500-8957","affiliations":[{"raw_affiliation_string":"Waseda University, Tokyo, Japan","institution_ids":["https://openalex.org/I150744194"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013781881","display_name":"Niko Moritz","orcid":"https://orcid.org/0000-0003-4352-8115"},"institutions":[{"id":"https://openalex.org/I4210111288","display_name":"Meta (United Kingdom)","ror":"https://ror.org/020ye1821","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210111288","https://openalex.org/I4210114444"]},{"id":"https://openalex.org/I4210159266","display_name":"Mitsubishi Electric (United States)","ror":"https://ror.org/053jnhe44","country_code":"US","type":"company","lineage":["https://openalex.org/I1306287861","https://openalex.org/I4210133125","https://openalex.org/I4210159266"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Niko Moritz","raw_affiliation_strings":["Mitsubishi Electric Research Laboratories (MERL), Cambridge, MA, USA","Meta AI, London, U.K"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Mitsubishi Electric Research Laboratories (MERL), Cambridge, MA, USA","institution_ids":["https://openalex.org/I4210159266"]},{"raw_affiliation_string":"Meta AI, London, U.K","institution_ids":["https://openalex.org/I4210111288"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076453358","display_name":"Jonathan Le Roux","orcid":"https://orcid.org/0000-0002-3451-171X"},"institutions":[{"id":"https://openalex.org/I4210159266","display_name":"Mitsubishi Electric (United States)","ror":"https://ror.org/053jnhe44","country_code":"US","type":"company","lineage":["https://openalex.org/I1306287861","https://openalex.org/I4210133125","https://openalex.org/I4210159266"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jonathan Le Roux","raw_affiliation_strings":["Mitsubishi Electric Research Laboratories (MERL), Cambridge, MA, USA"],"raw_orcid":"https://orcid.org/0000-0002-3451-171X","affiliations":[{"raw_affiliation_string":"Mitsubishi Electric Research Laboratories (MERL), Cambridge, MA, USA","institution_ids":["https://openalex.org/I4210159266"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5087554069","display_name":"Takaaki Hori","orcid":"https://orcid.org/0000-0003-4560-8039"},"institutions":[{"id":"https://openalex.org/I4210153776","display_name":"Apple (United States)","ror":"https://ror.org/059hsda18","country_code":"US","type":"company","lineage":["https://openalex.org/I4210153776"]},{"id":"https://openalex.org/I4210159266","display_name":"Mitsubishi Electric (United States)","ror":"https://ror.org/053jnhe44","country_code":"US","type":"company","lineage":["https://openalex.org/I1306287861","https://openalex.org/I4210133125","https://openalex.org/I4210159266"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Takaaki Hori","raw_affiliation_strings":["Mitsubishi Electric Research Laboratories (MERL), Cambridge, MA, USA","Apple, Inc., Cambridge, MA, USA"],"raw_orcid":"https://orcid.org/0000-0003-4560-8039","affiliations":[{"raw_affiliation_string":"Mitsubishi Electric Research Laboratories (MERL), Cambridge, MA, USA","institution_ids":["https://openalex.org/I4210159266"]},{"raw_affiliation_string":"Apple, Inc., Cambridge, MA, USA","institution_ids":["https://openalex.org/I4210153776"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":{"value":2195,"currency":"USD","value_usd":2195},"apc_paid":{"value":2195,"currency":"USD","value_usd":2195},"fwci":2.6901,"has_fulltext":true,"cited_by_count":24,"citation_normalized_percentile":{"value":0.91077422,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":93,"max":99},"biblio":{"volume":"16","issue":"6","first_page":"1424","last_page":"1438"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9987000226974487,"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":0.9987000226974487,"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":0.9936000108718872,"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":0.9842000007629395,"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":0.7472500801086426},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6515523195266724},{"id":"https://openalex.org/keywords/connectionism","display_name":"Connectionism","score":0.5403760671615601},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4869893491268158},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4865097403526306},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.48146024346351624},{"id":"https://openalex.org/keywords/labeled-data","display_name":"Labeled data","score":0.44404298067092896},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4216653108596802},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3926456570625305},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.35345739126205444},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.33094656467437744}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7472500801086426},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6515523195266724},{"id":"https://openalex.org/C8521452","wikidata":"https://www.wikidata.org/wiki/Q203790","display_name":"Connectionism","level":3,"score":0.5403760671615601},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4869893491268158},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4865097403526306},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.48146024346351624},{"id":"https://openalex.org/C2776145971","wikidata":"https://www.wikidata.org/wiki/Q30673951","display_name":"Labeled data","level":2,"score":0.44404298067092896},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4216653108596802},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3926456570625305},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35345739126205444},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.33094656467437744},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/jstsp.2022.3195367","is_oa":true,"landing_page_url":"https://doi.org/10.1109/jstsp.2022.3195367","pdf_url":"https://ieeexplore.ieee.org/ielx7/4200690/9923627/09846970.pdf","source":{"id":"https://openalex.org/S42167783","display_name":"IEEE Journal of Selected Topics in Signal Processing","issn_l":"1932-4553","issn":["1932-4553","1941-0484"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Journal of Selected Topics in Signal Processing","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1109/jstsp.2022.3195367","is_oa":true,"landing_page_url":"https://doi.org/10.1109/jstsp.2022.3195367","pdf_url":"https://ieeexplore.ieee.org/ielx7/4200690/9923627/09846970.pdf","source":{"id":"https://openalex.org/S42167783","display_name":"IEEE Journal of Selected Topics in Signal Processing","issn_l":"1932-4553","issn":["1932-4553","1941-0484"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Journal of Selected Topics in Signal Processing","raw_type":"journal-article"},"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.7799999713897705}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4289824098.pdf","grobid_xml":"https://content.openalex.org/works/W4289824098.grobid-xml"},"referenced_works_count":96,"referenced_works":["https://openalex.org/W82886505","https://openalex.org/W1494198834","https://openalex.org/W1710082047","https://openalex.org/W1821462560","https://openalex.org/W2111316763","https://openalex.org/W2127141656","https://openalex.org/W2143612262","https://openalex.org/W2144499799","https://openalex.org/W2160815625","https://openalex.org/W2327501763","https://openalex.org/W2407080277","https://openalex.org/W2502312327","https://openalex.org/W2531409750","https://openalex.org/W2577366047","https://openalex.org/W2627092829","https://openalex.org/W2799473636","https://openalex.org/W2889213362","https://openalex.org/W2892008152","https://openalex.org/W2892009249","https://openalex.org/W2896457183","https://openalex.org/W2913851961","https://openalex.org/W2936774411","https://openalex.org/W2940322076","https://openalex.org/W2948981900","https://openalex.org/W2962699523","https://openalex.org/W2962780374","https://openalex.org/W2962824709","https://openalex.org/W2963739817","https://openalex.org/W2963979492","https://openalex.org/W2964110616","https://openalex.org/W2972818416","https://openalex.org/W2973040747","https://openalex.org/W2973049979","https://openalex.org/W2995181338","https://openalex.org/W3007073761","https://openalex.org/W3007328579","https://openalex.org/W3015265920","https://openalex.org/W3015356564","https://openalex.org/W3015522062","https://openalex.org/W3015537910","https://openalex.org/W3015737168","https://openalex.org/W3016011332","https://openalex.org/W3026041220","https://openalex.org/W3035160371","https://openalex.org/W3035524453","https://openalex.org/W3046667470","https://openalex.org/W3093579165","https://openalex.org/W3095350795","https://openalex.org/W3096273170","https://openalex.org/W3096338464","https://openalex.org/W3097747488","https://openalex.org/W3097777922","https://openalex.org/W3101648800","https://openalex.org/W3103005696","https://openalex.org/W3141100132","https://openalex.org/W3156828761","https://openalex.org/W3160766462","https://openalex.org/W3160799772","https://openalex.org/W3161143478","https://openalex.org/W3162833755","https://openalex.org/W3163464943","https://openalex.org/W3163793923","https://openalex.org/W3193590960","https://openalex.org/W3197223534","https://openalex.org/W3197916665","https://openalex.org/W3198098585","https://openalex.org/W3205405669","https://openalex.org/W3206573929","https://openalex.org/W4210606444","https://openalex.org/W4225755266","https://openalex.org/W4250482878","https://openalex.org/W4286912221","https://openalex.org/W6623517193","https://openalex.org/W6631190155","https://openalex.org/W6631362777","https://openalex.org/W6638523607","https://openalex.org/W6638667902","https://openalex.org/W6675365184","https://openalex.org/W6679434410","https://openalex.org/W6679436768","https://openalex.org/W6724804524","https://openalex.org/W6733590821","https://openalex.org/W6733814495","https://openalex.org/W6739901393","https://openalex.org/W6762242920","https://openalex.org/W6763608318","https://openalex.org/W6764051988","https://openalex.org/W6765939562","https://openalex.org/W6768222176","https://openalex.org/W6769196770","https://openalex.org/W6774314701","https://openalex.org/W6779326418","https://openalex.org/W6780218876","https://openalex.org/W6780226713","https://openalex.org/W6784614252","https://openalex.org/W6840182953"],"related_works":["https://openalex.org/W4380075502","https://openalex.org/W4223943233","https://openalex.org/W4312200629","https://openalex.org/W4360585206","https://openalex.org/W4364306694","https://openalex.org/W4380086463","https://openalex.org/W4225161397","https://openalex.org/W3014300295","https://openalex.org/W3164822677","https://openalex.org/W2795261237"],"abstract_inverted_index":{"End-to-end":[0],"automatic":[1],"speech":[2],"recognition":[3],"(ASR)":[4],"has":[5,65],"become":[6],"a":[7,19,35,61,73,89,100,169],"popular":[8],"alternative":[9],"to":[10,45,69,121,168,221,227],"traditional":[11],"module-based":[12],"systems,":[13],"simplifying":[14],"the":[15,26,53,112,126,129,141,148,161,196,201,212],"model-building":[16],"process":[17],"with":[18,78,181],"single":[20],"deep":[21],"neural":[22],"network":[23],"architecture.":[24],"However,":[25],"training":[27,54,154,202],"of":[28,38,99,140,163,184,214],"end-to-end":[29,70],"ASR":[30,153],"systems":[31],"is":[32,43,76,119],"generally":[33],"data-hungry:":[34],"large":[36],"amount":[37],"labeled":[39,58],"data":[40,157,185],"(speech-text":[41],"pairs)":[42],"necessary":[44],"learn":[46,106],"direct":[47],"speech-to-text":[48],"conversion":[49],"effectively.":[50],"To":[51],"make":[52],"less":[55],"dependent":[56],"on":[57,125,155,177,203,235],"data,":[59],"pseudo-labeling,":[60],"semi-supervised":[62,95,179,232],"learning":[63],"approach,":[64],"been":[66],"successfully":[67],"introduced":[68],"ASR,":[71],"where":[72],"seed":[74,197],"model":[75,118,134,143,173,198],"self-trained":[77],"pseudo-labels":[79,123],"generated":[80,124],"from":[81,107],"unlabeled":[82,156,204],"(speech-only)":[83],"data.":[84,205],"Here,":[85],"we":[86,207],"proposemomentum":[87],"pseudo-labeling(MPL),":[88],"simple":[90],"yet":[91],"effective":[92],"strategy":[93],"for":[94],"ASR.":[96],"MPL":[97,167,193],"consists":[98],"pair":[101],"ofonlineandofflinemodels":[102],"that":[103,192],"interact":[104],"and":[105,174,216],"each":[108],"other,":[109],"inspired":[110],"by":[111,128,158,199],"mean":[113],"teacher":[114],"method.":[115],"The":[116,132,145,189],"online":[117,142],"trained":[120],"predict":[122],"fly":[127],"offline":[130,133],"model.":[131],"maintains":[135],"an":[136,217],"exponential":[137],"moving":[138],"average":[139],"parameters.":[144],"interaction":[146],"between":[147],"two":[149],"models":[150],"allows":[151],"better":[152,228],"continuously":[159],"improving":[160],"quality":[162],"pseudo-labels.":[164],"We":[165],"apply":[166],"connectionist":[170],"temporal":[171],"classification-based":[172],"evaluate":[175],"it":[176],"various":[178],"scenarios":[180],"varying":[182],"amounts":[183],"or":[186],"domain":[187],"mismatch.":[188],"results":[190],"demonstrate":[191],"significantly":[194],"improves":[195],"stabilizing":[200],"Moreover,":[206],"present":[208],"additional":[209],"techniques,":[210],"e.g.,":[211],"use":[213],"Conformer":[215],"external":[218],"language":[219],"model,":[220],"further":[222],"enhance":[223],"MPL,":[224],"which":[225],"leads":[226],"performance":[229],"than":[230],"other":[231],"methods":[233],"based":[234],"pseudo-labeling.":[236]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":7},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":10},{"year":2022,"cited_by_count":2}],"updated_date":"2026-08-28T12:50:07.497085","created_date":"2025-10-10T00:00:00"}