{"id":"https://openalex.org/W4391021561","doi":"https://doi.org/10.1109/asru57964.2023.10389725","title":"Pseudo-Label Based Supervised Contrastive Loss for Robust Speech Representations","display_name":"Pseudo-Label Based Supervised Contrastive Loss for Robust Speech Representations","publication_year":2023,"publication_date":"2023-12-16","ids":{"openalex":"https://openalex.org/W4391021561","doi":"https://doi.org/10.1109/asru57964.2023.10389725"},"language":"en","primary_location":{"id":"doi:10.1109/asru57964.2023.10389725","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/asru57964.2023.10389725","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","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/A5053022933","display_name":"Varun Krishna","orcid":null},"institutions":[{"id":"https://openalex.org/I59270414","display_name":"Indian Institute of Science Bangalore","ror":"https://ror.org/04dese585","country_code":"IN","type":"education","lineage":["https://openalex.org/I59270414"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Varun Krishna","raw_affiliation_strings":["Learning and Extraction of Acoustic Patterns (LEAP) Lab, Indian Institute of Science,Bangalore","Learning and Extraction of Acoustic Patterns (LEAP) Lab, Indian Institute of Science, Bangalore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Learning and Extraction of Acoustic Patterns (LEAP) Lab, Indian Institute of Science,Bangalore","institution_ids":["https://openalex.org/I59270414"]},{"raw_affiliation_string":"Learning and Extraction of Acoustic Patterns (LEAP) Lab, Indian Institute of Science, Bangalore","institution_ids":["https://openalex.org/I59270414"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5002536077","display_name":"Sriram Ganapathy","orcid":"https://orcid.org/0000-0002-5779-9066"},"institutions":[{"id":"https://openalex.org/I59270414","display_name":"Indian Institute of Science Bangalore","ror":"https://ror.org/04dese585","country_code":"IN","type":"education","lineage":["https://openalex.org/I59270414"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Sriram Ganapathy","raw_affiliation_strings":["Learning and Extraction of Acoustic Patterns (LEAP) Lab, Indian Institute of Science,Bangalore","Learning and Extraction of Acoustic Patterns (LEAP) Lab, Indian Institute of Science, Bangalore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Learning and Extraction of Acoustic Patterns (LEAP) Lab, Indian Institute of Science,Bangalore","institution_ids":["https://openalex.org/I59270414"]},{"raw_affiliation_string":"Learning and Extraction of Acoustic Patterns (LEAP) Lab, Indian Institute of Science, Bangalore","institution_ids":["https://openalex.org/I59270414"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.19066443,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"33","issue":null,"first_page":"1","last_page":"8"},"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":0.9990000128746033,"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/T11309","display_name":"Music and Audio Processing","score":0.9979000091552734,"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":0.8116341829299927},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5754654407501221},{"id":"https://openalex.org/keywords/cross-entropy","display_name":"Cross entropy","score":0.5728700160980225},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5207157731056213},{"id":"https://openalex.org/keywords/cluster-analysis","display_name":"Cluster analysis","score":0.5049319863319397},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.47321122884750366},{"id":"https://openalex.org/keywords/lexical-analysis","display_name":"Lexical analysis","score":0.4505676031112671},{"id":"https://openalex.org/keywords/timit","display_name":"TIMIT","score":0.4326586127281189},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.426662415266037},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.4149319529533386},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.411138117313385},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3301457166671753},{"id":"https://openalex.org/keywords/hidden-markov-model","display_name":"Hidden Markov model","score":0.2995898723602295}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8116341829299927},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5754654407501221},{"id":"https://openalex.org/C167981619","wikidata":"https://www.wikidata.org/wiki/Q1685498","display_name":"Cross entropy","level":3,"score":0.5728700160980225},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5207157731056213},{"id":"https://openalex.org/C73555534","wikidata":"https://www.wikidata.org/wiki/Q622825","display_name":"Cluster analysis","level":2,"score":0.5049319863319397},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.47321122884750366},{"id":"https://openalex.org/C176982825","wikidata":"https://www.wikidata.org/wiki/Q835922","display_name":"Lexical analysis","level":2,"score":0.4505676031112671},{"id":"https://openalex.org/C2778724510","wikidata":"https://www.wikidata.org/wiki/Q7670405","display_name":"TIMIT","level":3,"score":0.4326586127281189},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.426662415266037},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.4149319529533386},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.411138117313385},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3301457166671753},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.2995898723602295},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","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/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/asru57964.2023.10389725","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/asru57964.2023.10389725","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2023 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"},{"id":"pmh:oai::84557","is_oa":false,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4306401429","display_name":"ePrints@IISc (Indian Institute of Science)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I59270414","host_organization_name":"Indian Institute of Science Bangalore","host_organization_lineage":["https://openalex.org/I59270414"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"","raw_type":"Conference Paper"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":58,"referenced_works":["https://openalex.org/W108866686","https://openalex.org/W1494198834","https://openalex.org/W1635512741","https://openalex.org/W1647671624","https://openalex.org/W2030931454","https://openalex.org/W2085487226","https://openalex.org/W2129068307","https://openalex.org/W2219249508","https://openalex.org/W2395899413","https://openalex.org/W2431080869","https://openalex.org/W2726515241","https://openalex.org/W2741692265","https://openalex.org/W2842511635","https://openalex.org/W2896457183","https://openalex.org/W2973157397","https://openalex.org/W2979476256","https://openalex.org/W2996728628","https://openalex.org/W3006926732","https://openalex.org/W3016011332","https://openalex.org/W3036601975","https://openalex.org/W3041561163","https://openalex.org/W3113328489","https://openalex.org/W3144810982","https://openalex.org/W3156636935","https://openalex.org/W3159481202","https://openalex.org/W3163596720","https://openalex.org/W3197349023","https://openalex.org/W3198608154","https://openalex.org/W3198858531","https://openalex.org/W3200253633","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W4226033575","https://openalex.org/W4281492411","https://openalex.org/W4283026156","https://openalex.org/W4287591426","https://openalex.org/W4287812705","https://openalex.org/W4292825791","https://openalex.org/W4297808394","https://openalex.org/W4306672449","https://openalex.org/W4312766345","https://openalex.org/W4319862416","https://openalex.org/W4385822676","https://openalex.org/W4389317789","https://openalex.org/W6678975374","https://openalex.org/W6688816777","https://openalex.org/W6717772578","https://openalex.org/W6755207826","https://openalex.org/W6769196770","https://openalex.org/W6774314701","https://openalex.org/W6776700526","https://openalex.org/W6780218876","https://openalex.org/W6780483730","https://openalex.org/W6786696081","https://openalex.org/W6810673746","https://openalex.org/W6838461927","https://openalex.org/W6839738141","https://openalex.org/W6846365670"],"related_works":["https://openalex.org/W80423236","https://openalex.org/W3164669818","https://openalex.org/W4386014872","https://openalex.org/W1573546415","https://openalex.org/W1847536016","https://openalex.org/W4361193986","https://openalex.org/W3149094754","https://openalex.org/W2148703997","https://openalex.org/W4366851046","https://openalex.org/W2994927414"],"abstract_inverted_index":{"The":[0,60,87],"self":[1],"supervised":[2],"learning":[3,104],"(SSL)":[4],"of":[5,42,66,75,83,110,146,152],"speech,":[6],"with":[7,95],"discrete":[8],"tokenization":[9],"(pseudo-labels),":[10],"while":[11,79],"illustrating":[12],"performance":[13,168],"improvements":[14,169],"in":[15,22,102],"low-resource":[16],"speech":[17],"recognition,":[18],"has":[19],"faced":[20],"challenges":[21],"achieving":[23],"context":[24,161],"invariant":[25],"and":[26,132],"noise":[27],"robust":[28],"representations.":[29],"In":[30],"this":[31],"paper,":[32],"we":[33,165],"propose":[34],"a":[35,64,84],"self-supervised":[36,103],"framework":[37,143],"based":[38],"on":[39,149,159,174],"contrastive":[40],"loss":[41,90,113],"the":[43,55,69,73,76,81,96,108,111,140,160,175,183],"pseudo-labels,":[44],"obtained":[45],"from":[46,125],"an":[47],"offline":[48],"k-means":[49],"quantizer":[50,130],"(tokenizer).":[51],"We":[52,106],"refer":[53],"to":[54,71],"proposed":[56,88,141],"setting":[57],"as":[58,156,158,180,182],"pseudo-con.":[59],"pseudo-con":[61,89,112,142],"loss,":[62,99],"within":[63],"batch":[65],"training,":[67],"allows":[68],"model":[70],"cluster":[72],"instances":[74,82],"same":[77],"pseudo-label":[78],"separating":[80],"different":[85],"pseudo-label.":[86],"can":[91],"also":[92],"be":[93],"combined":[94],"cross":[97],"entropy":[98],"commonly":[100],"used":[101],"schemes.":[105],"demonstrate":[107],"effectiveness":[109],"applied":[114],"for":[115],"various":[116,150],"SSL":[117,172],"techniques,":[118],"like":[119],"hidden":[120,133],"unit":[121,134],"bidirectional":[122],"encoder":[123],"representations":[124],"transformers":[126],"(HuBERT),":[127],"best":[128],"random":[129],"(BEST-RQ)":[131],"clustering":[135],"(HUC).":[136],"Our":[137],"evaluations":[138],"using":[139],"achieves":[144],"state":[145],"art":[147],"results":[148],"sub-tasks":[151],"ZeroSpeech":[153],"2021":[154],"challenge":[155],"well":[157,181],"invariance":[162],"benchmarks.":[163],"Further,":[164],"show":[166],"significant":[167],"over":[170],"existing":[171],"approaches":[173],"TIMIT":[176],"phoneme":[177],"recognition":[178],"task":[179],"Librispeech":[184],"(100h)":[185],"ASR":[186],"experiments.":[187]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}