{"id":"https://openalex.org/W3200129129","doi":"https://doi.org/10.1109/icassp43922.2022.9747432","title":"Performance-Efficiency Trade-Offs in Unsupervised Pre-Training for Speech Recognition","display_name":"Performance-Efficiency Trade-Offs in Unsupervised Pre-Training for Speech Recognition","publication_year":2022,"publication_date":"2022-04-27","ids":{"openalex":"https://openalex.org/W3200129129","doi":"https://doi.org/10.1109/icassp43922.2022.9747432","mag":"3200129129"},"language":"en","primary_location":{"id":"doi:10.1109/icassp43922.2022.9747432","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9747432","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":"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/A5074393700","display_name":"Felix Wu","orcid":null},"institutions":[{"id":"https://openalex.org/I1283576576","display_name":"NetApp (United States)","ror":"https://ror.org/05c4cm338","country_code":"US","type":"company","lineage":["https://openalex.org/I1283576576"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Felix Wu","raw_affiliation_strings":["ASAPP Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ASAPP Inc","institution_ids":["https://openalex.org/I1283576576"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052377938","display_name":"Kwangyoun Kim","orcid":null},"institutions":[{"id":"https://openalex.org/I1283576576","display_name":"NetApp (United States)","ror":"https://ror.org/05c4cm338","country_code":"US","type":"company","lineage":["https://openalex.org/I1283576576"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kwangyoun Kim","raw_affiliation_strings":["ASAPP Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ASAPP Inc","institution_ids":["https://openalex.org/I1283576576"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101875418","display_name":"Jing Pan","orcid":"https://orcid.org/0000-0002-5178-2247"},"institutions":[{"id":"https://openalex.org/I1283576576","display_name":"NetApp (United States)","ror":"https://ror.org/05c4cm338","country_code":"US","type":"company","lineage":["https://openalex.org/I1283576576"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jing Pan","raw_affiliation_strings":["ASAPP Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ASAPP Inc","institution_ids":["https://openalex.org/I1283576576"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106859201","display_name":"Kyu J. Han","orcid":null},"institutions":[{"id":"https://openalex.org/I1283576576","display_name":"NetApp (United States)","ror":"https://ror.org/05c4cm338","country_code":"US","type":"company","lineage":["https://openalex.org/I1283576576"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kyu J. Han","raw_affiliation_strings":["ASAPP Inc"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ASAPP Inc","institution_ids":["https://openalex.org/I1283576576"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006380227","display_name":"Kilian Q. Weinberger","orcid":null},"institutions":[{"id":"https://openalex.org/I1283576576","display_name":"NetApp (United States)","ror":"https://ror.org/05c4cm338","country_code":"US","type":"company","lineage":["https://openalex.org/I1283576576"]},{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kilian Q. Weinberger","raw_affiliation_strings":["ASAPP Inc","Cornell University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ASAPP Inc","institution_ids":["https://openalex.org/I1283576576"]},{"raw_affiliation_string":"Cornell University","institution_ids":["https://openalex.org/I205783295"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5047026141","display_name":"Yoav Artzi","orcid":"https://orcid.org/0000-0002-4605-6144"},"institutions":[{"id":"https://openalex.org/I1283576576","display_name":"NetApp (United States)","ror":"https://ror.org/05c4cm338","country_code":"US","type":"company","lineage":["https://openalex.org/I1283576576"]},{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yoav Artzi","raw_affiliation_strings":["ASAPP Inc","Cornell University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ASAPP Inc","institution_ids":["https://openalex.org/I1283576576"]},{"raw_affiliation_string":"Cornell University","institution_ids":["https://openalex.org/I205783295"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":27,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"7667","last_page":"7671"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9998000264167786,"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":0.9998000264167786,"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":0.9994999766349792,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9994000196456909,"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":0.7745700478553772},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7294737100601196},{"id":"https://openalex.org/keywords/word-error-rate","display_name":"Word error rate","score":0.7201286554336548},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6366324424743652},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5987379550933838},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.5920307636260986},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.5500409603118896},{"id":"https://openalex.org/keywords/variety","display_name":"Variety (cybernetics)","score":0.47618380188941956},{"id":"https://openalex.org/keywords/acoustic-model","display_name":"Acoustic model","score":0.475232869386673},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.47177839279174805},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.44995787739753723},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.42096057534217834},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.33542758226394653},{"id":"https://openalex.org/keywords/speech-processing","display_name":"Speech processing","score":0.2867829203605652},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.12618675827980042},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.11346703767776489}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7745700478553772},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7294737100601196},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.7201286554336548},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6366324424743652},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5987379550933838},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.5920307636260986},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.5500409603118896},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.47618380188941956},{"id":"https://openalex.org/C155635449","wikidata":"https://www.wikidata.org/wiki/Q4674699","display_name":"Acoustic model","level":3,"score":0.475232869386673},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.47177839279174805},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.44995787739753723},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42096057534217834},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.33542758226394653},{"id":"https://openalex.org/C61328038","wikidata":"https://www.wikidata.org/wiki/Q3358061","display_name":"Speech processing","level":2,"score":0.2867829203605652},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.12618675827980042},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.11346703767776489},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0},{"id":"https://openalex.org/C120665830","wikidata":"https://www.wikidata.org/wiki/Q14620","display_name":"Optics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp43922.2022.9747432","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp43922.2022.9747432","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":102,"referenced_works":["https://openalex.org/W10548402","https://openalex.org/W179875071","https://openalex.org/W1494198834","https://openalex.org/W1524333225","https://openalex.org/W1828163288","https://openalex.org/W1836465849","https://openalex.org/W1991133427","https://openalex.org/W2127141656","https://openalex.org/W2143612262","https://openalex.org/W2155273149","https://openalex.org/W2160815625","https://openalex.org/W2193413348","https://openalex.org/W2296701362","https://openalex.org/W2327501763","https://openalex.org/W2331143823","https://openalex.org/W2345968833","https://openalex.org/W2520160253","https://openalex.org/W2547875792","https://openalex.org/W2622263826","https://openalex.org/W2626778328","https://openalex.org/W2795935804","https://openalex.org/W2799473636","https://openalex.org/W2804648901","https://openalex.org/W2842511635","https://openalex.org/W2892009249","https://openalex.org/W2896457183","https://openalex.org/W2898700502","https://openalex.org/W2933138175","https://openalex.org/W2946948417","https://openalex.org/W2949117887","https://openalex.org/W2952509486","https://openalex.org/W2963341956","https://openalex.org/W2963925437","https://openalex.org/W2963970792","https://openalex.org/W2964121744","https://openalex.org/W2965373594","https://openalex.org/W2970971581","https://openalex.org/W2973049979","https://openalex.org/W2975381464","https://openalex.org/W2979476256","https://openalex.org/W2981363336","https://openalex.org/W2982413405","https://openalex.org/W2987283559","https://openalex.org/W2988736778","https://openalex.org/W2995181338","https://openalex.org/W3001279689","https://openalex.org/W3004728855","https://openalex.org/W3005680577","https://openalex.org/W3021469861","https://openalex.org/W3025165719","https://openalex.org/W3026408381","https://openalex.org/W3027083471","https://openalex.org/W3033187248","https://openalex.org/W3035060554","https://openalex.org/W3035524453","https://openalex.org/W3036601975","https://openalex.org/W3037057938","https://openalex.org/W3046727238","https://openalex.org/W3093533780","https://openalex.org/W3093579165","https://openalex.org/W3097777922","https://openalex.org/W3099782249","https://openalex.org/W3101648800","https://openalex.org/W3119308075","https://openalex.org/W3122890974","https://openalex.org/W3126565544","https://openalex.org/W3127866760","https://openalex.org/W3134206242","https://openalex.org/W3135828102","https://openalex.org/W3144073776","https://openalex.org/W3144173820","https://openalex.org/W3153287399","https://openalex.org/W3157916917","https://openalex.org/W3160525311","https://openalex.org/W3160799772","https://openalex.org/W3162391496","https://openalex.org/W3164279099","https://openalex.org/W3197642003","https://openalex.org/W3198299542","https://openalex.org/W3198429080","https://openalex.org/W3198771897","https://openalex.org/W4297808394","https://openalex.org/W4322614701","https://openalex.org/W4385245566","https://openalex.org/W6629717138","https://openalex.org/W6638667902","https://openalex.org/W6697501834","https://openalex.org/W6727336983","https://openalex.org/W6729448088","https://openalex.org/W6739901393","https://openalex.org/W6755207826","https://openalex.org/W6755862295","https://openalex.org/W6766673545","https://openalex.org/W6770717842","https://openalex.org/W6774314701","https://openalex.org/W6779068807","https://openalex.org/W6779326418","https://openalex.org/W6779919476","https://openalex.org/W6780218876","https://openalex.org/W6784614252","https://openalex.org/W6784637704","https://openalex.org/W6795346631"],"related_works":["https://openalex.org/W2097707447","https://openalex.org/W2944691285","https://openalex.org/W3081187864","https://openalex.org/W4380605396","https://openalex.org/W2803306015","https://openalex.org/W3133352777","https://openalex.org/W151018310","https://openalex.org/W2784059283","https://openalex.org/W4319779560","https://openalex.org/W2008737763"],"abstract_inverted_index":{"This":[0],"paper":[1],"is":[2],"a":[3,50,64,80,89,98],"study":[4],"of":[5,66],"performance-efficiency":[6],"trade-offs":[7],"in":[8,93],"pre-trained":[9,51],"models":[10],"for":[11],"automatic":[12],"speech":[13],"recognition":[14],"(ASR).":[15],"We":[16],"focus":[17],"on":[18,76],"wav2vec":[19,86],"2.0,":[20,87],"and":[21,32,44,60],"formalize":[22],"several":[23],"architecture":[24,53],"designs":[25],"that":[26],"influence":[27],"both":[28,58],"the":[29,72],"model":[30,52,111],"performance":[31,59],"its":[33],"efficiency.":[34],"Putting":[35],"together":[36],"all":[37],"our":[38],"observations,":[39],"we":[40],"introduce":[41],"SEW-D":[42,78],"(Squeezed":[43],"Efficient":[45],"Wav2vec":[46],"with":[47,54,88],"Disentangled":[48],"Attention),":[49],"significant":[55],"improvements":[56],"along":[57],"efficiency":[61],"dimensions":[62],"across":[63,109],"variety":[65],"training":[67],"setups.":[68],"For":[69],"example,":[70],"under":[71],"100h-960h":[73],"semi-supervised":[74],"setup":[75],"LibriSpeech,":[77],"achieves":[79],"1.9x":[81],"inference":[82,100],"speedup":[83],"compared":[84],"to":[85],"13.5%":[90],"relative":[91],"reduction":[92],"word":[94,104],"error":[95,105],"rate.":[96],"With":[97],"similar":[99],"time,":[101],"SEW":[102],"reduces":[103],"rate":[106],"by":[107],"25\u201350%":[108],"different":[110],"sizes.":[112]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":8},{"year":2023,"cited_by_count":12},{"year":2022,"cited_by_count":2}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}