{"id":"https://openalex.org/W4224934179","doi":"https://doi.org/10.1109/icassp43922.2022.9747022","title":"Improving Self-Supervised Learning for Speech Recognition with Intermediate Layer Supervision","display_name":"Improving Self-Supervised Learning for Speech Recognition with Intermediate Layer Supervision","publication_year":2022,"publication_date":"2022-04-27","ids":{"openalex":"https://openalex.org/W4224934179","doi":"https://doi.org/10.1109/icassp43922.2022.9747022"},"language":"en","primary_location":{"id":"doi:10.1109/icassp43922.2022.9747022","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9747022","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/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"]},{"id":"https://openalex.org/I205237279","display_name":"Nankai University","ror":"https://ror.org/01y1kjr75","country_code":"CN","type":"education","lineage":["https://openalex.org/I205237279"]}],"countries":["CN","FI"],"is_corresponding":true,"raw_author_name":"Chengyi Wang","raw_affiliation_strings":["NanKai University,China","Microsoft Corporation","NanKai University, China"],"affiliations":[{"raw_affiliation_string":"NanKai University,China","institution_ids":["https://openalex.org/I205237279"]},{"raw_affiliation_string":"Microsoft Corporation","institution_ids":["https://openalex.org/I4210105678"]},{"raw_affiliation_string":"NanKai University, China","institution_ids":["https://openalex.org/I205237279"]}]},{"author_position":"middle","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"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079533447","display_name":"Sanyuan Chen","orcid":"https://orcid.org/0000-0002-3082-6052"},"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":"Sanyuan Chen","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/A5101635405","display_name":"Shujie Liu","orcid":"https://orcid.org/0009-0008-0785-8882"},"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":"Shujie Liu","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/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":"last","author":{"id":"https://openalex.org/A5013188461","display_name":"Zhenglu Yang","orcid":"https://orcid.org/0000-0001-9528-965X"},"institutions":[{"id":"https://openalex.org/I205237279","display_name":"Nankai University","ror":"https://ror.org/01y1kjr75","country_code":"CN","type":"education","lineage":["https://openalex.org/I205237279"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhenglu Yang","raw_affiliation_strings":["NanKai University,China","NanKai University, China"],"affiliations":[{"raw_affiliation_string":"NanKai University,China","institution_ids":["https://openalex.org/I205237279"]},{"raw_affiliation_string":"NanKai University, China","institution_ids":["https://openalex.org/I205237279"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5101863385"],"corresponding_institution_ids":["https://openalex.org/I205237279","https://openalex.org/I4210105678"],"apc_list":null,"apc_paid":null,"fwci":1.8788,"has_fulltext":false,"cited_by_count":19,"citation_normalized_percentile":{"value":0.87517888,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"7092","last_page":"7096"},"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/T11309","display_name":"Music and Audio Processing","score":0.9991000294685364,"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/T10860","display_name":"Speech and Audio Processing","score":0.9958999752998352,"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":0.8423799276351929},{"id":"https://openalex.org/keywords/intuition","display_name":"Intuition","score":0.6354027390480042},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5561498403549194},{"id":"https://openalex.org/keywords/encode","display_name":"ENCODE","score":0.5396873950958252},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.522231936454773},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5124734044075012},{"id":"https://openalex.org/keywords/layer","display_name":"Layer (electronics)","score":0.5120754837989807},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.4867120087146759},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4814562201499939},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.43099308013916016},{"id":"https://openalex.org/keywords/acoustic-model","display_name":"Acoustic model","score":0.4265354871749878},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.34613606333732605}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8423799276351929},{"id":"https://openalex.org/C132010649","wikidata":"https://www.wikidata.org/wiki/Q189222","display_name":"Intuition","level":2,"score":0.6354027390480042},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5561498403549194},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.5396873950958252},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.522231936454773},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5124734044075012},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.5120754837989807},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.4867120087146759},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4814562201499939},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.43099308013916016},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.4265354871749878},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.34613606333732605},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp43922.2022.9747022","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9747022","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","id":"https://metadata.un.org/sdg/4","score":0.7699999809265137}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":38,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2842511635","https://openalex.org/W2896457183","https://openalex.org/W2972943112","https://openalex.org/W2973049979","https://openalex.org/W2973157397","https://openalex.org/W2979476256","https://openalex.org/W2982223350","https://openalex.org/W2995181338","https://openalex.org/W3003875258","https://openalex.org/W3015213852","https://openalex.org/W3016011332","https://openalex.org/W3036601975","https://openalex.org/W3041561163","https://openalex.org/W3093579165","https://openalex.org/W3112034174","https://openalex.org/W3144810982","https://openalex.org/W3167207712","https://openalex.org/W3197580070","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W4226033575","https://openalex.org/W4287173589","https://openalex.org/W4288089799","https://openalex.org/W4297808394","https://openalex.org/W4385245566","https://openalex.org/W6739901393","https://openalex.org/W6755207826","https://openalex.org/W6769196770","https://openalex.org/W6769593479","https://openalex.org/W6769627184","https://openalex.org/W6780218876","https://openalex.org/W6784614252","https://openalex.org/W6786669483","https://openalex.org/W6788335241","https://openalex.org/W6795952400","https://openalex.org/W6803164887","https://openalex.org/W6947929050"],"related_works":["https://openalex.org/W2594897229","https://openalex.org/W2151348424","https://openalex.org/W4221142855","https://openalex.org/W2050138804","https://openalex.org/W4398173524","https://openalex.org/W4290708361","https://openalex.org/W2129812225","https://openalex.org/W26527944","https://openalex.org/W2523799048","https://openalex.org/W2155620340"],"abstract_inverted_index":{"Recently,":[0],"pioneer":[1],"work":[2],"finds":[3],"that":[4,98],"self-supervised":[5],"pre-training":[6],"methods":[7],"can":[8],"improve":[9],"multiple":[10],"downstream":[11],"speech":[12,40],"tasks,":[13],"because":[14],"the":[15,32,39,48,71,89,114,125,146],"model":[16,49,72,117,130,159],"utilizes":[17],"bottom":[18,126],"layers":[19,26,127],"to":[20,27,52,73],"learn":[21],"speaker-related":[22],"information":[23,55,77],"and":[24,144,158],"top":[25],"encode":[28],"content-related":[29],"information.":[30],"Since":[31],"network":[33],"capacity":[34],"is":[35,50,139],"limited,":[36],"we":[37,60],"believe":[38],"recognition":[41],"performance":[42],"could":[43],"be":[44],"further":[45],"improved":[46],"if":[47],"dedicated":[51],"audio":[53],"content":[54,76],"learning.":[56],"To":[57],"this":[58],"end,":[59],"propose":[61],"Intermediate":[62],"Layer":[63],"Supervision":[64],"for":[65,119,151],"Self-Supervised":[66],"Learning":[67],"(ILS-SSL),":[68],"which":[69,104,138],"forces":[70],"concentrate":[74],"on":[75,88,93],"as":[78,80],"much":[79],"possible":[81],"by":[82],"adding":[83],"an":[84],"additional":[85],"SSL":[86],"loss":[87],"intermediate":[90],"layers.":[91],"Experiments":[92],"LibriSpeech":[94],"test-other":[95],"set":[96],"show":[97],"our":[99,129,142,149,156],"method":[100,150],"outperforms":[101],"HuBERT":[102],"significantly,":[103],"achieves":[105],"a":[106,132],"23.5%/11.6%":[107],"relative":[108],"word":[109],"error":[110],"rate":[111],"reduction":[112],"in":[113],"w/o":[115],"language":[116],"setting":[118],"Base/Large":[120],"models.":[121],"Detailed":[122],"analysis":[123],"shows":[124],"of":[128,148],"have":[131],"better":[133],"correlation":[134],"with":[135,141],"phonetic":[136],"units,":[137],"consistent":[140],"intuition":[143],"explains":[145],"success":[147],"ASR.":[152],"We":[153],"will":[154],"release":[155],"code":[157],"at":[160],"https://github.com/microsoft/UniSpeech.":[161]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":6},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":4}],"updated_date":"2026-04-03T22:45:19.894376","created_date":"2025-10-10T00:00:00"}