{"id":"https://openalex.org/W3205533980","doi":"https://doi.org/10.1109/icassp43922.2022.9746929","title":"Wav2vec-Switch: Contrastive Learning from Original-Noisy Speech Pairs for Robust Speech Recognition","display_name":"Wav2vec-Switch: Contrastive Learning from Original-Noisy Speech Pairs for Robust Speech Recognition","publication_year":2022,"publication_date":"2022-04-27","ids":{"openalex":"https://openalex.org/W3205533980","doi":"https://doi.org/10.1109/icassp43922.2022.9746929","mag":"3205533980"},"language":"en","primary_location":{"id":"doi:10.1109/icassp43922.2022.9746929","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9746929","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":"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/A5100377935","display_name":"Yiming Wang","orcid":"https://orcid.org/0000-0002-5932-4371"},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":true,"raw_author_name":"Yiming Wang","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100365053","display_name":"Jinyu Li","orcid":"https://orcid.org/0000-0002-1089-9748"},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Jinyu Li","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101497069","display_name":"Heming Wang","orcid":"https://orcid.org/0000-0003-3861-0624"},"institutions":[{"id":"https://openalex.org/I52357470","display_name":"The Ohio State University","ror":"https://ror.org/00rs6vg23","country_code":"US","type":"education","lineage":["https://openalex.org/I52357470"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Heming Wang","raw_affiliation_strings":["The Ohio State University,USA","The Ohio State University, USA"],"affiliations":[{"raw_affiliation_string":"The Ohio State University,USA","institution_ids":["https://openalex.org/I52357470"]},{"raw_affiliation_string":"The Ohio State University, USA","institution_ids":["https://openalex.org/I52357470"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100342006","display_name":"Yao Qian","orcid":"https://orcid.org/0000-0003-1855-9630"},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Yao Qian","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101863385","display_name":"Chengyi Wang","orcid":"https://orcid.org/0000-0002-6780-9299"},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Chengyi Wang","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101709477","display_name":"Yu Wu","orcid":"https://orcid.org/0000-0002-5715-3011"},"institutions":[{"id":"https://openalex.org/I4210105678","display_name":"Microsoft (Finland)","ror":"https://ror.org/01nehjf29","country_code":"FI","type":"company","lineage":["https://openalex.org/I1290206253","https://openalex.org/I4210105678"]}],"countries":["FI"],"is_corresponding":false,"raw_author_name":"Yu Wu","raw_affiliation_strings":["Microsoft Corporation"],"affiliations":[{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5100377935"],"corresponding_institution_ids":["https://openalex.org/I4210105678"],"apc_list":null,"apc_paid":null,"fwci":5.3069,"has_fulltext":false,"cited_by_count":54,"citation_normalized_percentile":{"value":0.96622313,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"7097","last_page":"7101"},"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/T10860","display_name":"Speech and Audio Processing","score":0.9997000098228455,"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":0.998199999332428,"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/robustness","display_name":"Robustness (evolution)","score":0.811733603477478},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7794987559318542},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.7478350400924683},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.623255729675293},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.6053644418716431},{"id":"https://openalex.org/keywords/voice-activity-detection","display_name":"Voice activity detection","score":0.46765297651290894},{"id":"https://openalex.org/keywords/noise-measurement","display_name":"Noise measurement","score":0.4595213830471039},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4493626654148102},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.44693124294281006},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.4067225158214569},{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.3810839354991913},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3495182394981384},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.09520766139030457},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.07251638174057007}],"concepts":[{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.811733603477478},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7794987559318542},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.7478350400924683},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.623255729675293},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.6053644418716431},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.46765297651290894},{"id":"https://openalex.org/C29265498","wikidata":"https://www.wikidata.org/wiki/Q7047719","display_name":"Noise measurement","level":3,"score":0.4595213830471039},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4493626654148102},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.44693124294281006},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.4067225158214569},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.3810839354991913},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3495182394981384},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.09520766139030457},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.07251638174057007},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp43922.2022.9746929","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9746929","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":[{"display_name":"Quality Education","score":0.550000011920929,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":46,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2024490156","https://openalex.org/W2124509324","https://openalex.org/W2127141656","https://openalex.org/W2219249508","https://openalex.org/W2399016933","https://openalex.org/W2547875792","https://openalex.org/W2559260703","https://openalex.org/W2680270903","https://openalex.org/W2802248956","https://openalex.org/W2842511635","https://openalex.org/W2933138175","https://openalex.org/W2962892438","https://openalex.org/W2962894366","https://openalex.org/W2963403868","https://openalex.org/W2963522845","https://openalex.org/W2973049979","https://openalex.org/W2979476256","https://openalex.org/W2982223350","https://openalex.org/W2996383576","https://openalex.org/W2998616931","https://openalex.org/W3008762051","https://openalex.org/W3008912312","https://openalex.org/W3015213852","https://openalex.org/W3015265920","https://openalex.org/W3015312544","https://openalex.org/W3016011332","https://openalex.org/W3032514799","https://openalex.org/W3036601975","https://openalex.org/W3093579165","https://openalex.org/W3096710170","https://openalex.org/W3099782249","https://openalex.org/W3160799772","https://openalex.org/W3167207712","https://openalex.org/W3197411683","https://openalex.org/W4226033575","https://openalex.org/W4297808394","https://openalex.org/W4385245566","https://openalex.org/W6688816777","https://openalex.org/W6729448088","https://openalex.org/W6739901393","https://openalex.org/W6769196770","https://openalex.org/W6773243159","https://openalex.org/W6780218876","https://openalex.org/W6784614252","https://openalex.org/W6788335241"],"related_works":["https://openalex.org/W642007152","https://openalex.org/W4287867034","https://openalex.org/W3108403339","https://openalex.org/W2401827384","https://openalex.org/W2355290951","https://openalex.org/W2069501481","https://openalex.org/W2162712524","https://openalex.org/W2142924612","https://openalex.org/W2052688117","https://openalex.org/W3005783148"],"abstract_inverted_index":{"The":[0],"goal":[1],"of":[2,21,59,89,99,139],"self-supervised":[3],"learning":[4,82],"(SSL)":[5],"for":[6,24,41,113,183],"automatic":[7],"speech":[8,15,23,60,68,94,197],"recognition":[9],"(ASR)":[10],"is":[11,39],"to":[12,52,78,109,121,172],"learn":[13,122],"good":[14],"representations":[16,58,88],"from":[17],"a":[18,50,173,179],"large":[19],"amount":[20],"unlabeled":[22],"the":[25,72,79,86,90,107,114,137,152,160],"downstream":[26],"ASR":[27],"task.":[28],"However,":[29],"most":[30],"SSL":[31],"frameworks":[32],"do":[33],"not":[34],"consider":[35],"noise":[36,54,126],"robustness":[37,55],"which":[38],"crucial":[40],"real-world":[42],"applications.":[43],"In":[44,76],"this":[45],"paper":[46],"we":[47,65,84],"propose":[48],"wav2vec-Switch,":[49],"method":[51],"encode":[53],"into":[56,71],"contextualized":[57,123],"via":[61],"contrastive":[62,81],"learning.":[63],"Specifically,":[64],"feed":[66],"original-noisy":[67],"pairs":[69],"simultaneously":[70],"wav2vec":[73],"2.0":[74],"network.":[75],"addition":[77],"existing":[80],"task,":[83],"switch":[85],"quantized":[87],"original":[91,115,161],"and":[92,116,132,163],"noisy":[93,117,134,154,169],"as":[95],"additional":[96],"prediction":[97],"targets":[98],"each":[100],"other.":[101],"By":[102],"doing":[103],"this,":[104],"it":[105,142],"enforces":[106],"network":[108],"have":[110],"consistent":[111],"predictions":[112],"speech,":[118],"thus":[119],"allows":[120],"representation":[124],"with":[125,178,195],"robustness.":[127],"Our":[128,185],"experiments":[129],"on":[130,151,159,165,187],"synthe-sized":[131],"real":[133,167],"data":[135,156,170,174],"show":[136],"effectiveness":[138],"our":[140],"method:":[141],"achieves":[143],"2.9\u20134.9%":[144],"relative":[145],"word":[146],"error":[147],"rate":[148],"(WER)":[149],"reduction":[150],"synthesized":[153],"LibriSpeech":[155],"without":[157],"deterioration":[158],"data,":[162],"5.7%":[164],"CHiME-4":[166,188],"1-channel":[168],"compared":[171],"augmentation":[175],"baseline":[176],"even":[177,192],"strong":[180],"language":[181],"model":[182],"decoding.":[184],"results":[186],"can":[189],"match":[190],"or":[191],"surpass":[193],"those":[194],"well-designed":[196],"enhancement":[198],"components.":[199]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":20},{"year":2023,"cited_by_count":15},{"year":2022,"cited_by_count":5},{"year":2021,"cited_by_count":1}],"updated_date":"2026-01-20T17:24:06.736184","created_date":"2025-10-10T00:00:00"}