{"id":"https://openalex.org/W7167679779","doi":"https://doi.org/10.21437/odyssey.2026-48","title":"Flow-Enhanced Language Embeddings for Robust Language Recognition","display_name":"Flow-Enhanced Language Embeddings for Robust Language Recognition","publication_year":2026,"publication_date":"2026-06-23","ids":{"openalex":"https://openalex.org/W7167679779","doi":"https://doi.org/10.21437/odyssey.2026-48"},"language":"en","primary_location":{"id":"doi:10.21437/odyssey.2026-48","doi":"https://doi.org/10.21437/odyssey.2026-48","is_oa":true,"landing_page_url":"https://doi.org/10.21437/odyssey.2026-48","pdf_url":"https://www.isca-archive.org/odyssey_2026/cao26_odyssey.pdf","source":{"id":"https://openalex.org/S7407086350","display_name":"The Speaker and Language Recognition Workshop (Odyssey)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Odyssey 2026","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://www.isca-archive.org/odyssey_2026/cao26_odyssey.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5014505688","display_name":"Tianyu Cao","orcid":"https://orcid.org/0000-0002-2661-7004","observed_orcids":["https://orcid.org/0000-0002-2661-7004"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tianyu Cao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126968265","display_name":"Laureano Moro Velazquez","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Laureano Moro Velazquez","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057010207","display_name":"Jes\u00fas Villalba","orcid":"https://orcid.org/0000-0001-9459-8426","observed_orcids":["https://orcid.org/0000-0001-9459-8426"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jesus Villalba","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040324682","display_name":"Thomas Thebaud","orcid":"https://orcid.org/0000-0001-8953-7872","observed_orcids":["https://orcid.org/0000-0001-8953-7872"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Thomas Thebaud","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5050632169","display_name":"Najim Dehak","orcid":"https://orcid.org/0000-0002-4489-5753","observed_orcids":["https://orcid.org/0000-0002-4489-5753"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Najim Dehak","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.4041885,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"328","last_page":"335"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9980000257492065,"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":0.9980000257492065,"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":0.0008999999845400453,"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":0.0005000000237487257,"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/language-identification","display_name":"language identification","score":0.9739999771118164},{"id":"https://openalex.org/keywords/flow-matching","display_name":"flow matching","score":0.7549999952316284},{"id":"https://openalex.org/keywords/embedding-refinement","display_name":"embedding refinement","score":0.5879999995231628},{"id":"https://openalex.org/keywords/domain-mismatch","display_name":"domain mismatch","score":0.7860000133514404},{"id":"https://openalex.org/keywords/far-field-speech","display_name":"far-field speech","score":0.5600000023841858}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6248999834060669},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5335000157356262},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.49559998512268066},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.38190001249313354},{"id":"https://openalex.org/C129792486","wikidata":"https://www.wikidata.org/wiki/Q1050419","display_name":"Language identification","level":3,"score":0.35179999470710754},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.3296000063419342},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3206000030040741},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.29339998960494995},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.29330000281333923},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.27790001034736633},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2549999952316284}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/odyssey.2026-48","doi":"https://doi.org/10.21437/odyssey.2026-48","is_oa":true,"landing_page_url":"https://doi.org/10.21437/odyssey.2026-48","pdf_url":"https://www.isca-archive.org/odyssey_2026/cao26_odyssey.pdf","source":{"id":"https://openalex.org/S7407086350","display_name":"The Speaker and Language Recognition Workshop (Odyssey)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Odyssey 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.21437/odyssey.2026-48","doi":"https://doi.org/10.21437/odyssey.2026-48","is_oa":true,"landing_page_url":"https://doi.org/10.21437/odyssey.2026-48","pdf_url":"https://www.isca-archive.org/odyssey_2026/cao26_odyssey.pdf","source":{"id":"https://openalex.org/S7407086350","display_name":"The Speaker and Language Recognition Workshop (Odyssey)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Odyssey 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7167679779.pdf","grobid_xml":"https://content.openalex.org/works/W7167679779.grobid-xml"},"referenced_works_count":15,"referenced_works":["https://openalex.org/W2972359262","https://openalex.org/W3037932933","https://openalex.org/W3095410713","https://openalex.org/W3096408984","https://openalex.org/W3106807794","https://openalex.org/W4303647933","https://openalex.org/W4319050105","https://openalex.org/W4376456759","https://openalex.org/W4400877621","https://openalex.org/W4402112025","https://openalex.org/W4402112435","https://openalex.org/W4415433133","https://openalex.org/W7140173642","https://openalex.org/W7155069372","https://openalex.org/W7196954125"],"related_works":["https://openalex.org/W4390905769","https://openalex.org/W4297841657","https://openalex.org/W3150131481","https://openalex.org/W7203972556","https://openalex.org/W7123357104","https://openalex.org/W2487408737","https://openalex.org/W2888957357","https://openalex.org/W7222069158","https://openalex.org/W2054115175","https://openalex.org/W7122729662"],"abstract_inverted_index":{"Language":[0],"identification":[1],"(LID)":[2],"systems":[3],"are":[4],"widely":[5],"used":[6],"in":[7,20],"speech":[8],"analytics,":[9],"multilingual":[10],"routing,":[11],"and":[12,34,64,68,108],"spoken-language":[13],"interfaces,":[14],"but":[15],"their":[16,65],"performance":[17],"often":[18],"drops":[19],"real-world":[21],"conditions":[22,118],"because":[23],"of":[24],"domain":[25],"mismatch,":[26],"including":[27],"background":[28],"noise,":[29],"reverberation,":[30],"channel":[31],"effects,":[32],"codecs,":[33],"far-field":[35],"recording.We":[36],"propose":[37],"an":[38],"embedding-level":[39],"generative":[40],"enhancement":[41],"method":[42,100],"based":[43],"on":[44],"flow":[45,92],"matching":[46],"that":[47,73],"refines":[48],"language":[49,110],"embeddings":[50,60,76],"from":[51,61],"a":[52,70,95],"pre-trained":[53],"LID":[54,106,114],"model.During":[55],"training,":[56],"we":[57],"build":[58],"paired":[59],"clean":[62,78],"utterances":[63],"distorted":[66],"versions,":[67],"learn":[69],"continuous-time":[71],"transformation":[72],"maps":[74],"corrupted":[75],"toward":[77],"ones":[79],"while":[80,112],"preserving":[81],"language-discriminative":[82],"structure.At":[83],"inference,":[84],"any":[85],"embedding":[86],"is":[87],"passed":[88],"through":[89],"the":[90,105],"learned":[91],"to":[93,104],"obtain":[94],"refined,":[96],"more":[97],"robust":[98],"representation.The":[99],"requires":[101],"no":[102,109],"changes":[103],"pipeline":[107],"labels,":[111],"improving":[113],"accuracy":[115],"under":[116],"challenging":[117],"without":[119],"hurting":[120],"standard":[121],"performance.":[122]},"counts_by_year":[],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2026-07-09T00:00:00"}