{"id":"https://openalex.org/W4385484924","doi":"https://doi.org/10.1109/icasspw59220.2023.10193042","title":"Fine-Tuning Strategies for Faster Inference Using Speech Self-Supervised Models: A Comparative Study","display_name":"Fine-Tuning Strategies for Faster Inference Using Speech Self-Supervised Models: A Comparative Study","publication_year":2023,"publication_date":"2023-06-04","ids":{"openalex":"https://openalex.org/W4385484924","doi":"https://doi.org/10.1109/icasspw59220.2023.10193042"},"language":"en","primary_location":{"id":"doi:10.1109/icasspw59220.2023.10193042","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icasspw59220.2023.10193042","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)","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/A5040954517","display_name":"Salah Zaiem","orcid":null},"institutions":[{"id":"https://openalex.org/I12356871","display_name":"T\u00e9l\u00e9com Paris","ror":"https://ror.org/01naq7912","country_code":"FR","type":"education","lineage":["https://openalex.org/I12356871","https://openalex.org/I205703379","https://openalex.org/I4210145102"]}],"countries":["FR"],"is_corresponding":true,"raw_author_name":"Salah Zaiem","raw_affiliation_strings":["Telecom Paris,Palaiseau,France","Telecom Paris, Palaiseau, France"],"affiliations":[{"raw_affiliation_string":"Telecom Paris,Palaiseau,France","institution_ids":["https://openalex.org/I12356871"]},{"raw_affiliation_string":"Telecom Paris, Palaiseau, France","institution_ids":["https://openalex.org/I12356871"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071809848","display_name":"Robin Algayres","orcid":null},"institutions":[{"id":"https://openalex.org/I29607241","display_name":"\u00c9cole Normale Sup\u00e9rieure - PSL","ror":"https://ror.org/05a0dhs15","country_code":"FR","type":"other","lineage":["https://openalex.org/I2746051580","https://openalex.org/I29607241"]},{"id":"https://openalex.org/I154526488","display_name":"Inserm","ror":"https://ror.org/02vjkv261","country_code":"FR","type":"government","lineage":["https://openalex.org/I154526488"]},{"id":"https://openalex.org/I1326498283","display_name":"Institut national de recherche en sciences et technologies du num\u00e9rique","ror":"https://ror.org/02kvxyf05","country_code":"FR","type":"government","lineage":["https://openalex.org/I1326498283"]},{"id":"https://openalex.org/I2746051580","display_name":"Universit\u00e9 Paris Sciences et Lettres","ror":"https://ror.org/013cjyk83","country_code":"FR","type":"education","lineage":["https://openalex.org/I2746051580"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Robin Algayres","raw_affiliation_strings":["PSL Research University,ENS, INRIA, INSERM, UPEC,France","ENS, INRIA, INSERM, UPEC, PSL Research University, France"],"affiliations":[{"raw_affiliation_string":"PSL Research University,ENS, INRIA, INSERM, UPEC,France","institution_ids":["https://openalex.org/I154526488","https://openalex.org/I29607241","https://openalex.org/I1326498283"]},{"raw_affiliation_string":"ENS, INRIA, INSERM, UPEC, PSL Research University, France","institution_ids":["https://openalex.org/I2746051580","https://openalex.org/I154526488","https://openalex.org/I1326498283","https://openalex.org/I29607241"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089505434","display_name":"Titouan Parcollet","orcid":"https://orcid.org/0000-0003-0672-1346"},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Titouan Parcollet","raw_affiliation_strings":["Samsung AI Center,Cambridge,United-Kingdom","Samsung AI Center, Cambridge, United-Kingdom"],"affiliations":[{"raw_affiliation_string":"Samsung AI Center,Cambridge,United-Kingdom","institution_ids":["https://openalex.org/I4210117523"]},{"raw_affiliation_string":"Samsung AI Center, Cambridge, United-Kingdom","institution_ids":["https://openalex.org/I4210117523"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060031161","display_name":"Slim Essid","orcid":"https://orcid.org/0000-0002-0028-327X"},"institutions":[{"id":"https://openalex.org/I12356871","display_name":"T\u00e9l\u00e9com Paris","ror":"https://ror.org/01naq7912","country_code":"FR","type":"education","lineage":["https://openalex.org/I12356871","https://openalex.org/I205703379","https://openalex.org/I4210145102"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Slim Essid","raw_affiliation_strings":["Telecom Paris,Palaiseau,France","Telecom Paris, Palaiseau, France"],"affiliations":[{"raw_affiliation_string":"Telecom Paris,Palaiseau,France","institution_ids":["https://openalex.org/I12356871"]},{"raw_affiliation_string":"Telecom Paris, Palaiseau, France","institution_ids":["https://openalex.org/I12356871"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102962935","display_name":"Mirco Ravanelli","orcid":"https://orcid.org/0000-0001-8479-3121"},"institutions":[{"id":"https://openalex.org/I4210164802","display_name":"Mila - Quebec Artificial Intelligence Institute","ror":"https://ror.org/05c22rx21","country_code":"CA","type":"facility","lineage":["https://openalex.org/I4210164802"]},{"id":"https://openalex.org/I60158472","display_name":"Concordia University","ror":"https://ror.org/0420zvk78","country_code":"CA","type":"education","lineage":["https://openalex.org/I60158472"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Mirco Ravanelli","raw_affiliation_strings":["Mila-Quebec AI Institute, Université de Montréal, Concordia University,Canada"],"affiliations":[{"raw_affiliation_string":"Mila-Quebec AI Institute, Université de Montréal, Concordia University,Canada","institution_ids":["https://openalex.org/I4210164802","https://openalex.org/I60158472"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5040954517"],"corresponding_institution_ids":["https://openalex.org/I12356871"],"apc_list":null,"apc_paid":null,"fwci":1.9231,"has_fulltext":false,"cited_by_count":11,"citation_normalized_percentile":{"value":0.88665483,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"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":0.9990000128746033,"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":0.9984999895095825,"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":0.8294743299484253},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.7671098709106445},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7393327355384827},{"id":"https://openalex.org/keywords/upsampling","display_name":"Upsampling","score":0.7233169078826904},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5636671781539917},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.5591692328453064},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5466606020927429},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5200833678245544},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.49731186032295227},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.4015677273273468},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3419976532459259},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.1166052520275116}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8294743299484253},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.7671098709106445},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7393327355384827},{"id":"https://openalex.org/C110384440","wikidata":"https://www.wikidata.org/wiki/Q1143270","display_name":"Upsampling","level":3,"score":0.7233169078826904},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5636671781539917},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5591692328453064},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5466606020927429},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5200833678245544},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.49731186032295227},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4015677273273468},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3419976532459259},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.1166052520275116},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","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},{"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},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","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}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icasspw59220.2023.10193042","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icasspw59220.2023.10193042","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.6000000238418579,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":36,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W2024490156","https://openalex.org/W2134800885","https://openalex.org/W2145410271","https://openalex.org/W2963362078","https://openalex.org/W2975381464","https://openalex.org/W3036601975","https://openalex.org/W3044967013","https://openalex.org/W3096196861","https://openalex.org/W3136363192","https://openalex.org/W3167533889","https://openalex.org/W3203140070","https://openalex.org/W3209984917","https://openalex.org/W4221161761","https://openalex.org/W4224000641","https://openalex.org/W4224821750","https://openalex.org/W4226380987","https://openalex.org/W4289821214","https://openalex.org/W4296069141","https://openalex.org/W4297841557","https://openalex.org/W4311000453","https://openalex.org/W4313484417","https://openalex.org/W4319862442","https://openalex.org/W4319862480","https://openalex.org/W4372260433","https://openalex.org/W4372347449","https://openalex.org/W6656414902","https://openalex.org/W6679855610","https://openalex.org/W6768080748","https://openalex.org/W6780218876","https://openalex.org/W6802045652","https://openalex.org/W6803164887","https://openalex.org/W6810673746","https://openalex.org/W6810893203","https://openalex.org/W6848709529","https://openalex.org/W6849390140"],"related_works":["https://openalex.org/W2062399876","https://openalex.org/W2607795551","https://openalex.org/W3155117723","https://openalex.org/W1991429770","https://openalex.org/W1983892167","https://openalex.org/W2281134365","https://openalex.org/W4310746709","https://openalex.org/W4306309518","https://openalex.org/W4385574037","https://openalex.org/W4386075645"],"abstract_inverted_index":{"Self-supervised":[0],"learning":[1],"(SSL)":[2],"has":[3,20],"allowed":[4],"substantial":[5],"progress":[6],"in":[7,13,63,90,143],"Automatic":[8],"Speech":[9],"Recognition":[10],"(ASR)":[11],"performance":[12,39,86,115],"low-resource":[14],"settings.":[15],"In":[16],"this":[17],"context,":[18],"it":[19],"been":[21],"demonstrated":[22],"that":[23,51,96],"larger":[24],"self-supervised":[25],"feature":[26],"extractors":[27],"are":[28],"crucial":[29],"for":[30],"achieving":[31],"lower":[32],"downstream":[33,99],"ASR":[34,80],"error":[35],"rates.":[36],"Thus,":[37],"better":[38],"might":[40],"be":[41,53],"sanctioned":[42],"with":[43,112,125],"longer":[44],"inferences.":[45,70],"This":[46],"article":[47],"explores":[48],"different":[49],"approaches":[50],"may":[52],"deployed":[54],"during":[55],"the":[56,60,64,105,109,136,139],"fine-tuning":[57],"to":[58,68,78,141,148],"reduce":[59],"computations":[61,122],"needed":[62],"SSL":[65],"encoder,":[66],"leading":[67],"faster":[69],"We":[71],"adapt":[72],"a":[73,101],"number":[74],"of":[75,104,129,138],"existing":[76],"techniques":[77],"common":[79],"settings":[81],"and":[82,88,117],"benchmark":[83],"them,":[84],"displaying":[85],"drops":[87,116],"gains":[89],"inference":[91],"times.":[92],"Interestingly,":[93],"we":[94,134],"found":[95],"given":[97],"enough":[98],"data,":[100],"simple":[102],"downsampling":[103],"input":[106],"sequences":[107],"outperforms":[108],"other":[110],"methods":[111],"both":[113],"low":[114],"high":[118],"computational":[119],"savings,":[120],"reducing":[121],"by":[123],"61.3%":[124],"an":[126],"WER":[127],"increase":[128],"only":[130],"0.":[131],"81.":[132],"Finally,":[133],"analyze":[135],"robustness":[137],"comparison":[140],"changes":[142],"dataset":[144,149],"conditions,":[145],"revealing":[146],"sensitivity":[147],"size.":[150]},"counts_by_year":[{"year":2025,"cited_by_count":4},{"year":2024,"cited_by_count":5},{"year":2023,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}