{"id":"https://openalex.org/W4319862446","doi":"https://doi.org/10.1109/slt54892.2023.10023013","title":"On the Use of Semantically-Aligned Speech Representations for Spoken Language Understanding","display_name":"On the Use of Semantically-Aligned Speech Representations for Spoken Language Understanding","publication_year":2023,"publication_date":"2023-01-09","ids":{"openalex":"https://openalex.org/W4319862446","doi":"https://doi.org/10.1109/slt54892.2023.10023013"},"language":"en","primary_location":{"id":"doi:10.1109/slt54892.2023.10023013","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt54892.2023.10023013","pdf_url":null,"source":{"id":"https://openalex.org/S4363605953","display_name":"2022 IEEE Spoken Language Technology Workshop (SLT)","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":"2022 IEEE Spoken Language Technology Workshop (SLT)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://hal.science/hal-04155025v1/file/2210.05291.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5089134789","display_name":"Ga\u00eblle Laperri\u00e8re","orcid":null},"institutions":[{"id":"https://openalex.org/I4210119991","display_name":"Laboratoire Informatique d'Avignon","ror":"https://ror.org/02n399288","country_code":"FR","type":"facility","lineage":["https://openalex.org/I198415970","https://openalex.org/I4210119991"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Gaelle Laperriere","raw_affiliation_strings":["LIA - Avignon Université,France","LIA - Laboratoire Informatique d'Avignon (339 Chemin des Meinajaries Agroparc BP 1228 84911 Avignon cedex 9 - France)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"LIA - Avignon Université,France","institution_ids":["https://openalex.org/I4210119991"]},{"raw_affiliation_string":"LIA - Laboratoire Informatique d'Avignon (339 Chemin des Meinajaries Agroparc BP 1228 84911 Avignon cedex 9 - France)","institution_ids":["https://openalex.org/I4210119991"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060151795","display_name":"Valentin Pelloin","orcid":"https://orcid.org/0000-0002-1259-127X"},"institutions":[{"id":"https://openalex.org/I234216984","display_name":"Universit\u00e9 Nantes Angers Le Mans","ror":"https://ror.org/0406t3m57","country_code":"FR","type":"education","lineage":["https://openalex.org/I234216984"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Valentin Pelloin","raw_affiliation_strings":["LIUM - Le Mans Université,France"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"LIUM - Le Mans Université,France","institution_ids":["https://openalex.org/I234216984"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071001941","display_name":"Micka\u00ebl Rouvier","orcid":"https://orcid.org/0000-0003-3541-3385"},"institutions":[{"id":"https://openalex.org/I4210119991","display_name":"Laboratoire Informatique d'Avignon","ror":"https://ror.org/02n399288","country_code":"FR","type":"facility","lineage":["https://openalex.org/I198415970","https://openalex.org/I4210119991"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Mickael Rouvier","raw_affiliation_strings":["LIA - Avignon Université,France","LIA - Laboratoire Informatique d'Avignon (339 Chemin des Meinajaries Agroparc BP 1228 84911 Avignon cedex 9 - France)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"LIA - Avignon Université,France","institution_ids":["https://openalex.org/I4210119991"]},{"raw_affiliation_string":"LIA - Laboratoire Informatique d'Avignon (339 Chemin des Meinajaries Agroparc BP 1228 84911 Avignon cedex 9 - France)","institution_ids":["https://openalex.org/I4210119991"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061939508","display_name":"Themos Stafylakis","orcid":"https://orcid.org/0000-0002-9227-3588"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Themos Stafylakis","raw_affiliation_strings":["Omilia - Conversational Intelligence,Greece"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Omilia - Conversational Intelligence,Greece","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5081852717","display_name":"Yannick Est\u00e8ve","orcid":"https://orcid.org/0000-0002-3656-8883"},"institutions":[{"id":"https://openalex.org/I4210119991","display_name":"Laboratoire Informatique d'Avignon","ror":"https://ror.org/02n399288","country_code":"FR","type":"facility","lineage":["https://openalex.org/I198415970","https://openalex.org/I4210119991"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Yannick Esteve","raw_affiliation_strings":["LIA - Avignon Université,France","LIA - Laboratoire Informatique d'Avignon (339 Chemin des Meinajaries Agroparc BP 1228 84911 Avignon cedex 9 - France)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"LIA - Avignon Université,France","institution_ids":["https://openalex.org/I4210119991"]},{"raw_affiliation_string":"LIA - Laboratoire Informatique d'Avignon (339 Chemin des Meinajaries Agroparc BP 1228 84911 Avignon cedex 9 - France)","institution_ids":["https://openalex.org/I4210119991"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.6597,"has_fulltext":true,"cited_by_count":8,"citation_normalized_percentile":{"value":0.85776247,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"361","last_page":"368"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9997000098228455,"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":0.9997000098228455,"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":0.9994000196456909,"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"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9983000159263611,"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":0.8656399250030518},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.692575991153717},{"id":"https://openalex.org/keywords/utterance","display_name":"Utterance","score":0.638369083404541},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.6295192837715149},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5830737352371216},{"id":"https://openalex.org/keywords/sentence","display_name":"Sentence","score":0.5749642252922058},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.538203239440918},{"id":"https://openalex.org/keywords/software-portability","display_name":"Software portability","score":0.5174006223678589},{"id":"https://openalex.org/keywords/spoken-language","display_name":"Spoken language","score":0.5034200549125671},{"id":"https://openalex.org/keywords/clipping","display_name":"Clipping (morphology)","score":0.49331942200660706},{"id":"https://openalex.org/keywords/frame","display_name":"Frame (networking)","score":0.48781588673591614},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.48708510398864746},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.47851860523223877},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.40524643659591675},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.25508373975753784},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.19566449522972107}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8656399250030518},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.692575991153717},{"id":"https://openalex.org/C2775852435","wikidata":"https://www.wikidata.org/wiki/Q258403","display_name":"Utterance","level":2,"score":0.638369083404541},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6295192837715149},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5830737352371216},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.5749642252922058},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.538203239440918},{"id":"https://openalex.org/C63000827","wikidata":"https://www.wikidata.org/wiki/Q3080428","display_name":"Software portability","level":2,"score":0.5174006223678589},{"id":"https://openalex.org/C2776230583","wikidata":"https://www.wikidata.org/wiki/Q1322198","display_name":"Spoken language","level":2,"score":0.5034200549125671},{"id":"https://openalex.org/C2776848632","wikidata":"https://www.wikidata.org/wiki/Q853463","display_name":"Clipping (morphology)","level":2,"score":0.49331942200660706},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.48781588673591614},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.48708510398864746},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.47851860523223877},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.40524643659591675},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.25508373975753784},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.19566449522972107},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C76155785","wikidata":"https://www.wikidata.org/wiki/Q418","display_name":"Telecommunications","level":1,"score":0.0},{"id":"https://openalex.org/C94625758","wikidata":"https://www.wikidata.org/wiki/Q7163","display_name":"Politics","level":2,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/slt54892.2023.10023013","is_oa":false,"landing_page_url":"https://doi.org/10.1109/slt54892.2023.10023013","pdf_url":null,"source":{"id":"https://openalex.org/S4363605953","display_name":"2022 IEEE Spoken Language Technology Workshop (SLT)","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":"2022 IEEE Spoken Language Technology Workshop (SLT)","raw_type":"proceedings-article"},{"id":"pmh:oai:HAL:hal-04155025v1","is_oa":true,"landing_page_url":"https://hal.science/hal-04155025","pdf_url":"https://hal.science/hal-04155025v1/file/2210.05291.pdf","source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"2022 IEEE Spoken Language Technology Workshop (SLT), Jan 2023, Doha, Qatar. pp.361-368, ⟨10.1109/SLT54892.2023.10023013⟩","raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"pmh:oai:HAL:hal-04155025v1","is_oa":true,"landing_page_url":"https://hal.science/hal-04155025","pdf_url":"https://hal.science/hal-04155025v1/file/2210.05291.pdf","source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"2022 IEEE Spoken Language Technology Workshop (SLT), Jan 2023, Doha, Qatar. pp.361-368, ⟨10.1109/SLT54892.2023.10023013⟩","raw_type":"info:eu-repo/semantics/conferenceObject"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.7799999713897705,"display_name":"Quality Education"}],"awards":[{"id":"https://openalex.org/G1454328438","display_name":"Exchanges for SPEech ReseArch aNd TechnOlogies","funder_award_id":"101007666","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G1663281917","display_name":"Artificial Intelligence foR Semantically controlled SPEech UndeRstanding","funder_award_id":"ANR-19-CE23-0004","funder_id":"https://openalex.org/F4320320883","funder_display_name":"Agence Nationale de la Recherche"},{"id":"https://openalex.org/G3010527862","display_name":null,"funder_award_id":"ANR\u201019","funder_id":"https://openalex.org/F4320320883","funder_display_name":"Agence Nationale de la Recherche"},{"id":"https://openalex.org/G309983124","display_name":"Stream Learning for Multilingual Knowledge Transfer","funder_award_id":"957017","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G3908916144","display_name":null,"funder_award_id":"2022-AD011012565","funder_id":"https://openalex.org/F4320310145","funder_display_name":"Johns Hopkins University"},{"id":"https://openalex.org/G434337789","display_name":null,"funder_award_id":"HORIZON2020","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"}],"funders":[{"id":"https://openalex.org/F4320310145","display_name":"Johns Hopkins University","ror":"https://ror.org/00za53h95"},{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320320883","display_name":"Agence Nationale de la Recherche","ror":"https://ror.org/00rbzpz17"},{"id":"https://openalex.org/F4320326256","display_name":"Grand \u00c9quipement National De Calcul Intensif","ror":"https://ror.org/0010d1q40"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4319862446.pdf","grobid_xml":"https://content.openalex.org/works/W4319862446.grobid-xml"},"referenced_works_count":45,"referenced_works":["https://openalex.org/W204182981","https://openalex.org/W1649407914","https://openalex.org/W1922644628","https://openalex.org/W2249819665","https://openalex.org/W2524643686","https://openalex.org/W2614322402","https://openalex.org/W2894164357","https://openalex.org/W2914417638","https://openalex.org/W2963288440","https://openalex.org/W2970820898","https://openalex.org/W2972314145","https://openalex.org/W2972327934","https://openalex.org/W2982223350","https://openalex.org/W3005352739","https://openalex.org/W3015412890","https://openalex.org/W3015867372","https://openalex.org/W3015885816","https://openalex.org/W3016006013","https://openalex.org/W3016262400","https://openalex.org/W3036601975","https://openalex.org/W3039695075","https://openalex.org/W3041561163","https://openalex.org/W3080185952","https://openalex.org/W3095552229","https://openalex.org/W3098169801","https://openalex.org/W3122349645","https://openalex.org/W3148001440","https://openalex.org/W3177325166","https://openalex.org/W3197642003","https://openalex.org/W3203375531","https://openalex.org/W3206189675","https://openalex.org/W3209059054","https://openalex.org/W3209984917","https://openalex.org/W3213029956","https://openalex.org/W4213168938","https://openalex.org/W4221154554","https://openalex.org/W4225096077","https://openalex.org/W4226194717","https://openalex.org/W4286359908","https://openalex.org/W6677337689","https://openalex.org/W6755207826","https://openalex.org/W6778560980","https://openalex.org/W6780218876","https://openalex.org/W6804030475","https://openalex.org/W6810185941"],"related_works":["https://openalex.org/W107105315","https://openalex.org/W4367156293","https://openalex.org/W1584537303","https://openalex.org/W4388155270","https://openalex.org/W1872724644","https://openalex.org/W2750549761","https://openalex.org/W28826848","https://openalex.org/W2122272819","https://openalex.org/W2130894091","https://openalex.org/W2284708545"],"abstract_inverted_index":{"In":[0],"this":[1,96],"paper":[2],"we":[3,90],"examine":[4],"the":[5,19,33,36,47,56,67,70,75,81,84,92],"use":[6,68],"of":[7,69,74,86,94],"semantically-aligned":[8],"speech":[9,50],"representations":[10],"for":[11],"end-to-end":[12,87],"spoken":[13],"language":[14,99],"understanding":[15],"(SLU).":[16],"We":[17,64],"employ":[18],"recently-introduced":[20],"SAMU-XLSR":[21,71],"model,":[22],"which":[23],"is":[24],"designed":[25],"to":[26],"generate":[27],"a":[28],"single":[29],"embedding":[30],"that":[31,66],"captures":[32],"semantics":[34],"at":[35],"utterance":[37],"level,":[38],"semantically":[39],"aligned":[40],"across":[41],"different":[42],"languages.":[43],"This":[44],"model":[45,53,72,78,97],"combines":[46],"acoustic":[48],"frame-level":[49],"representation":[51],"learning":[52],"(XLS-R)":[54],"with":[55],"Language":[57],"Agnostic":[58],"BERT":[59],"Sentence":[60],"Embedding":[61],"(LaBSE)":[62],"model.":[63],"show":[65],"instead":[73],"initial":[76],"XLS-R":[77],"improves":[79],"significantly":[80],"performance":[82],"in":[83,101],"framework":[85],"SLU.":[88,102],"Finally,":[89],"present":[91],"benefits":[93],"using":[95],"towards":[98],"portability":[100]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":5}],"updated_date":"2026-08-26T07:47:46.906454","created_date":"2023-02-11T00:00:00"}