{"id":"https://openalex.org/W7152016174","doi":"https://doi.org/10.48550/arxiv.2604.05830","title":"\"OK Aura, Be Fair With Me\": Demographics-Agnostic Training for Bias Mitigation in Wake-up Word Detection","display_name":"\"OK Aura, Be Fair With Me\": Demographics-Agnostic Training for Bias Mitigation in Wake-up Word Detection","publication_year":2026,"publication_date":"2026-04-07","ids":{"openalex":"https://openalex.org/W7152016174","doi":"https://doi.org/10.48550/arxiv.2604.05830"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2604.05830","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2604.05830","pdf_url":"https://arxiv.org/pdf/2604.05830","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2604.05830","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5125724817","display_name":"Fernando L\u00f3pez","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"L\u00f3pez, Fernando","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133219484","display_name":"Paula Delgado-Santos","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Delgado-Santos, Paula","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133172634","display_name":"Pablo G\u00f3mez","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"G\u00f3mez, Pablo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079995562","display_name":"David Solans","orcid":"https://orcid.org/0000-0002-7288-3418","observed_orcids":["https://orcid.org/0000-0002-7288-3418"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Solans, David","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5109759968","display_name":"Jordi Palac\u00edn Luque","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luque, Jordi","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":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.9477999806404114,"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":0.9477999806404114,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.029400000348687172,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.006099999882280827,"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/demographic-bias","display_name":"demographic bias","score":0.8330000042915344},{"id":"https://openalex.org/keywords/data-augmentation","display_name":"data augmentation","score":0.7889999747276306},{"id":"https://openalex.org/keywords/knowledge-distillation","display_name":"knowledge distillation","score":0.9980000257492065},{"id":"https://openalex.org/keywords/sex-bias","display_name":"sex bias","score":0.5450000166893005},{"id":"https://openalex.org/keywords/age-bias","display_name":"age bias","score":0.9919999837875366},{"id":"https://openalex.org/keywords/accent-bias","display_name":"accent bias","score":0.9300000071525574}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6762999892234802},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.5346999764442444},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5340999960899353},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5339999794960022},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.5038999915122986},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.4993000030517578},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.4742000102996826},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.41830000281333923},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.37529999017715454},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.27869999408721924},{"id":"https://openalex.org/C2776756274","wikidata":"https://www.wikidata.org/wiki/Q181767","display_name":"Stress (linguistics)","level":2,"score":0.2784000039100647},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.2637999951839447},{"id":"https://openalex.org/C204201278","wikidata":"https://www.wikidata.org/wiki/Q1332614","display_name":"Voice activity detection","level":3,"score":0.2578999996185303}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2604.05830","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2604.05830","pdf_url":"https://arxiv.org/pdf/2604.05830","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2604.05830","doi":"https://doi.org/10.48550/arxiv.2604.05830","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.05830","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2604.05830","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2604.05830","pdf_url":"https://arxiv.org/pdf/2604.05830","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"score":0.8974000215530396,"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10"}],"sustainable_development_goals_aurora":[{"score":0.4511193633079529,"display_name":"Quality education","id":"https://metadata.un.org/sdg/4"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4415432682","https://openalex.org/W4404589661","https://openalex.org/W7125846376","https://openalex.org/W7214168487","https://openalex.org/W4296069263","https://openalex.org/W4385822300","https://openalex.org/W3159783145","https://openalex.org/W7148260033","https://openalex.org/W3089682003","https://openalex.org/W7160076446"],"abstract_inverted_index":{"Voice-based":[0],"interfaces":[1],"are":[2,63],"widely":[3],"used;":[4],"however,":[5],"achieving":[6],"fair":[7],"Wake-up":[8,150],"Word":[9,151],"detection":[10],"across":[11,107],"diverse":[12],"speaker":[13,109],"populations":[14],"remains":[15],"a":[16,55,102,118],"critical":[17],"challenge":[18],"due":[19],"to":[20,74,101,135],"persistent":[21],"demographic":[22,60,98],"biases.":[23],"This":[24,138],"study":[25,139],"evaluates":[26],"the":[27,47,114,136,141],"effectiveness":[28,142],"of":[29,39,82,113,122,143],"demographics-agnostic":[30,93],"training":[31,56,94],"techniques":[32,73,95,116],"in":[33,146,149],"mitigating":[34],"performance":[35,105],"disparities":[36],"among":[37],"speakers":[38],"varying":[40],"sex,":[41,125],"age,":[42,128],"and":[43,78,129],"accent.":[44],"We":[45,68],"utilize":[46],"OK":[48],"Aura":[49],"database":[50],"for":[51,65,124,127,131],"our":[52],"experiments,":[53],"employing":[54],"methodology":[57],"that":[58,91],"excludes":[59],"labels,":[61],"which":[62],"reserved":[64],"evaluation":[66],"purposes.":[67],"explore":[69],"(i)":[70],"data":[71],"augmentation":[72],"enhance":[75],"model":[76],"generalization":[77],"(ii)":[79],"knowledge":[80],"distillation":[81],"pre-trained":[83],"foundational":[84],"speech":[85],"models.":[86],"The":[87],"experimental":[88],"results":[89],"indicate":[90],"these":[92],"markedly":[96],"reduce":[97],"bias,":[99],"leading":[100],"more":[103],"equitable":[104],"profile":[106],"different":[108],"groups.":[110],"Specifically,":[111],"one":[112],"evaluated":[115],"achieves":[117],"Predictive":[119],"Disparity":[120],"reduction":[121],"39.94\\%":[123],"83.65\\%":[126],"40.48\\%":[130],"accent":[132],"when":[133],"compared":[134],"baseline.":[137],"highlights":[140],"label-agnostic":[144],"methodologies":[145],"fostering":[147],"fairness":[148],"detection.":[152]},"counts_by_year":[],"updated_date":"2026-10-31T12:09:11.096373","created_date":"2026-04-09T00:00:00"}