{"id":"https://openalex.org/W4405709818","doi":"https://doi.org/10.1109/iscslp63861.2024.10800400","title":"Ensemble Knowledge Distillation from Speech SSL Models Considering Inter-Teacher Differences","display_name":"Ensemble Knowledge Distillation from Speech SSL Models Considering Inter-Teacher Differences","publication_year":2024,"publication_date":"2024-11-07","ids":{"openalex":"https://openalex.org/W4405709818","doi":"https://doi.org/10.1109/iscslp63861.2024.10800400"},"language":"en","primary_location":{"id":"doi:10.1109/iscslp63861.2024.10800400","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iscslp63861.2024.10800400","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 14th International Symposium on Chinese Spoken Language Processing (ISCSLP)","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/A5016978905","display_name":"Pei-Jun Liao","orcid":null},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":true,"raw_author_name":"Pei-Jun Liao","raw_affiliation_strings":["College of Electrical Engineering and Computer Science, National Taiwan University"],"affiliations":[{"raw_affiliation_string":"College of Electrical Engineering and Computer Science, National Taiwan University","institution_ids":["https://openalex.org/I16733864"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040508737","display_name":"Hung-yi Lee","orcid":"https://orcid.org/0000-0002-9654-5747"},"institutions":[{"id":"https://openalex.org/I16733864","display_name":"National Taiwan University","ror":"https://ror.org/05bqach95","country_code":"TW","type":"education","lineage":["https://openalex.org/I16733864"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Hung-Yi Lee","raw_affiliation_strings":["College of Electrical Engineering and Computer Science, National Taiwan University"],"affiliations":[{"raw_affiliation_string":"College of Electrical Engineering and Computer Science, National Taiwan University","institution_ids":["https://openalex.org/I16733864"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5071214181","display_name":"Hsin\u2010Min Wang","orcid":"https://orcid.org/0000-0003-3599-5071"},"institutions":[{"id":"https://openalex.org/I4210098366","display_name":"Institute of Information Science, Academia Sinica","ror":"https://ror.org/00z83z196","country_code":"TW","type":"facility","lineage":["https://openalex.org/I4210098366","https://openalex.org/I84653119"]}],"countries":["TW"],"is_corresponding":false,"raw_author_name":"Hsin-Min Wang","raw_affiliation_strings":["Institute of Information Science, Academia Sinica"],"affiliations":[{"raw_affiliation_string":"Institute of Information Science, Academia Sinica","institution_ids":["https://openalex.org/I4210098366"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5016978905"],"corresponding_institution_ids":["https://openalex.org/I16733864"],"apc_list":null,"apc_paid":null,"fwci":0.3626,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.70169854,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"716","last_page":"720"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9818999767303467,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9818999767303467,"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/T12031","display_name":"Speech and dialogue systems","score":0.9714000225067139,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9194999933242798,"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":0.7307419180870056},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.6136008501052856},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.40911704301834106},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3689844012260437},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.07714179158210754}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7307419180870056},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.6136008501052856},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.40911704301834106},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3689844012260437},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.07714179158210754},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iscslp63861.2024.10800400","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iscslp63861.2024.10800400","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE 14th International Symposium on Chinese Spoken Language Processing (ISCSLP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.6800000071525574,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":37,"referenced_works":["https://openalex.org/W2146334809","https://openalex.org/W2407386500","https://openalex.org/W2508418541","https://openalex.org/W2972584841","https://openalex.org/W2995181338","https://openalex.org/W3093981631","https://openalex.org/W3148040514","https://openalex.org/W3161223924","https://openalex.org/W3197580070","https://openalex.org/W3203140070","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W3215615641","https://openalex.org/W4221146627","https://openalex.org/W4226033575","https://openalex.org/W4226103796","https://openalex.org/W4226302796","https://openalex.org/W4319862410","https://openalex.org/W4319862641","https://openalex.org/W4372270126","https://openalex.org/W4372270198","https://openalex.org/W4385823182","https://openalex.org/W6629717138","https://openalex.org/W6750615492","https://openalex.org/W6750665317","https://openalex.org/W6751425476","https://openalex.org/W6755207826","https://openalex.org/W6761176036","https://openalex.org/W6763701032","https://openalex.org/W6768021236","https://openalex.org/W6768851824","https://openalex.org/W6769196770","https://openalex.org/W6780218876","https://openalex.org/W6796457354","https://openalex.org/W6810007534","https://openalex.org/W6810905413","https://openalex.org/W6850513377"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W3204019825"],"abstract_inverted_index":{"In":[0],"the":[1,84,121,129],"realm":[2],"of":[3],"speech":[4,18,38],"processing,":[5],"Self-Supervised":[6],"Learning":[7],"(SSL)":[8],"models":[9,40,60,71,88,124],"such":[10,20],"as":[11,21,92,117],"HuBERT":[12],"are":[13,41,109],"widely":[14],"used":[15,53],"in":[16,94,128],"various":[17],"tasks":[19,127],"Automatic":[22],"Speech":[23],"Recognition":[24],"(ASR)":[25],"and":[26,31,44,89,106],"Spoken":[27],"Language":[28],"Understanding":[29],"(SLU),":[30],"have":[32,52],"achieved":[33],"impressive":[34],"results.":[35],"However,":[36],"these":[37],"SSL":[39,70],"often":[42],"large":[43],"require":[45],"significant":[46],"computational":[47],"resources.":[48],"Many":[49],"previous":[50],"studies":[51],"Knowledge":[54,65],"Distillation":[55,66],"(KD)":[56],"to":[57,98],"learn":[58],"compact":[59],"from":[61],"complex":[62],"models.":[63],"Ensemble":[64],"(EKD)":[67],"transfers":[68],"multiple":[69,78],"into":[72,138],"a":[73],"single":[74],"student":[75,95,123],"model":[76,96],"via":[77],"prediction":[79],"heads.":[80],"We":[81],"focus":[82],"on":[83,125],"differences":[85],"among":[86],"teacher":[87],"use":[90],"them":[91],"residuals":[93,137],"learning":[97,101],"achieve":[99],"additional":[100],"goals.":[102],"Residual":[103,107],"Prediction":[104],"Heads":[105],"Regularization":[108],"proposed.":[110],"Combining":[111],"RobustHuBERT":[112],"with":[113],"WavLM+":[114],"or":[115],"Data2vec-base":[116],"teachers,":[118],"we":[119],"evaluate":[120],"resulting":[122],"6":[126],"SUPERB":[130],"benchmark.":[131],"The":[132],"results":[133],"show":[134],"that":[135],"taking":[136],"account":[139],"improves":[140],"performance.":[141]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2025-12-27T23:08:20.325037","created_date":"2025-10-10T00:00:00"}