{"id":"https://openalex.org/W4408345595","doi":"https://doi.org/10.1109/icassp49660.2025.10887609","title":"Contrastive Knowledge Distillation for Embedding Refinement in Personalized Speech Enhancement","display_name":"Contrastive Knowledge Distillation for Embedding Refinement in Personalized Speech Enhancement","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408345595","doi":"https://doi.org/10.1109/icassp49660.2025.10887609"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10887609","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10887609","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 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/A5018626055","display_name":"T. Serre","orcid":"https://orcid.org/0000-0003-0846-0039"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Thomas Serre","raw_affiliation_strings":["Orosound,Paris,France"],"affiliations":[{"raw_affiliation_string":"Orosound,Paris,France","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082784886","display_name":"Mathieu Fontaine","orcid":"https://orcid.org/0000-0002-7657-6271"},"institutions":[{"id":"https://openalex.org/I4210165912","display_name":"Laboratoire Traitement et Communication de l\u2019Information","ror":"https://ror.org/057er4c39","country_code":"FR","type":"facility","lineage":["https://openalex.org/I12356871","https://openalex.org/I4210145102","https://openalex.org/I4210145102","https://openalex.org/I4210165912"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Mathieu Fontaine","raw_affiliation_strings":["Institut Polytechnique de Paris,LTCI, Télécom Paris,Palaiseau,France"],"affiliations":[{"raw_affiliation_string":"Institut Polytechnique de Paris,LTCI, Télécom Paris,Palaiseau,France","institution_ids":["https://openalex.org/I4210165912"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033795981","display_name":"\u00c9ric Benhaim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"\u00c9ric Benhaim","raw_affiliation_strings":["Orosound,Paris,France"],"affiliations":[{"raw_affiliation_string":"Orosound,Paris,France","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5060031161","display_name":"Slim Essid","orcid":"https://orcid.org/0000-0002-0028-327X"},"institutions":[{"id":"https://openalex.org/I4210165912","display_name":"Laboratoire Traitement et Communication de l\u2019Information","ror":"https://ror.org/057er4c39","country_code":"FR","type":"facility","lineage":["https://openalex.org/I12356871","https://openalex.org/I4210145102","https://openalex.org/I4210145102","https://openalex.org/I4210165912"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Slim Essid","raw_affiliation_strings":["Institut Polytechnique de Paris,LTCI, Télécom Paris,Palaiseau,France"],"affiliations":[{"raw_affiliation_string":"Institut Polytechnique de Paris,LTCI, Télécom Paris,Palaiseau,France","institution_ids":["https://openalex.org/I4210165912"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5018626055"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.561,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.78058241,"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":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9983000159263611,"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"}},"topics":[{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.9983000159263611,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9541000127792358,"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/T10057","display_name":"Face and Expression Recognition","score":0.9064000248908997,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.7019056081771851},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.6797747611999512},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.6500779390335083},{"id":"https://openalex.org/keywords/speech-enhancement","display_name":"Speech enhancement","score":0.5268287062644958},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.45056572556495667},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.43438422679901123},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.33281081914901733},{"id":"https://openalex.org/keywords/chemistry","display_name":"Chemistry","score":0.07901927828788757},{"id":"https://openalex.org/keywords/chromatography","display_name":"Chromatography","score":0.06576776504516602}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7019056081771851},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.6797747611999512},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.6500779390335083},{"id":"https://openalex.org/C2776182073","wikidata":"https://www.wikidata.org/wiki/Q7575395","display_name":"Speech enhancement","level":3,"score":0.5268287062644958},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.45056572556495667},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.43438422679901123},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33281081914901733},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.07901927828788757},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.06576776504516602},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp49660.2025.10887609","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10887609","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},{"id":"pmh:oai:HAL:hal-05467995v1","is_oa":false,"landing_page_url":"https://telecom-paris.hal.science/hal-05467995","pdf_url":null,"source":{"id":"https://openalex.org/S4406922461","display_name":"SPIRE - Sciences Po Institutional REpository","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":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Apr 2025, Hyderabad, France. pp.1-5, ⟨10.1109/icassp49660.2025.10887609⟩","raw_type":"Conference papers"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":29,"referenced_works":["https://openalex.org/W2150769028","https://openalex.org/W2696967604","https://openalex.org/W2784163702","https://openalex.org/W2890964092","https://openalex.org/W2951130829","https://openalex.org/W2973062255","https://openalex.org/W3015306128","https://openalex.org/W3015636705","https://openalex.org/W3016361963","https://openalex.org/W3024869864","https://openalex.org/W3097653961","https://openalex.org/W3126399577","https://openalex.org/W3161140524","https://openalex.org/W3161480375","https://openalex.org/W3162534564","https://openalex.org/W3196570692","https://openalex.org/W3198543387","https://openalex.org/W3205519288","https://openalex.org/W3206706278","https://openalex.org/W4224919629","https://openalex.org/W4226185896","https://openalex.org/W4312570973","https://openalex.org/W4372265824","https://openalex.org/W4372266552","https://openalex.org/W4392908343","https://openalex.org/W4401609309","https://openalex.org/W6688816777","https://openalex.org/W6771467084","https://openalex.org/W6771538394"],"related_works":["https://openalex.org/W2081900870","https://openalex.org/W2037549926","https://openalex.org/W2345479200","https://openalex.org/W2183306018","https://openalex.org/W2849310602","https://openalex.org/W3006008237","https://openalex.org/W2419146053","https://openalex.org/W4388890789","https://openalex.org/W3096184950","https://openalex.org/W4231424160"],"abstract_inverted_index":{"Personalized":[0],"speech":[1],"enhancement":[2,33,123],"(PSE)":[3],"has":[4],"shown":[5],"convincing":[6],"results":[7],"when":[8],"it":[9],"comes":[10],"to":[11,84,108],"extracting":[12],"a":[13,25,93,100,110,139],"known":[14],"target":[15,29,44,74],"voice":[16,30,45,75],"among":[17],"interfering":[18],"ones.":[19],"The":[20],"corresponding":[21],"systems":[22],"usually":[23],"incorporate":[24],"representation":[26],"of":[27,42,72,88],"the":[28,32,43,55,70,73,89,122,130],"within":[31,121],"system,":[34],"which":[35],"is":[36],"extracted":[37],"from":[38,113],"an":[39],"enrollment":[40],"clip":[41],"with":[46],"upstream":[47],"models.":[48],"Those":[49],"models":[50],"are":[51],"generally":[52],"heavy":[53],"as":[54],"speaker":[56,90,95],"embedding\u2019s":[57],"quality":[58],"directly":[59],"affects":[60],"PSE":[61,135],"performances.":[62],"Yet,":[63],"embeddings":[64],"generated":[65],"beforehand":[66],"cannot":[67],"account":[68],"for":[69],"variations":[71],"during":[76,125],"inference":[77,126],"time.":[78],"In":[79],"this":[80,119],"paper,":[81],"we":[82],"propose":[83],"perform":[85],"on-the-fly":[86],"refinement":[87],"embedding":[91],"using":[92],"tiny":[94],"encoder.":[96],"We":[97,116],"first":[98],"introduce":[99],"novel":[101],"contrastive":[102],"knowledge":[103],"distillation":[104],"methodology":[105],"in":[106],"order":[107],"train":[109],"150k-parameter":[111],"encoder":[112,120],"complex":[114],"embeddings.":[115],"then":[117],"use":[118],"system":[124],"and":[127],"show":[128],"that":[129],"proposed":[131],"method":[132],"greatly":[133],"improves":[134],"performances":[136],"while":[137],"maintaining":[138],"low":[140],"computational":[141],"load.":[142]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2025-12-28T23:10:05.387466","created_date":"2025-10-10T00:00:00"}