{"id":"https://openalex.org/W4372191701","doi":"https://doi.org/10.1109/icassp49357.2023.10094750","title":"Disentangled Training with Adversarial Examples for Robust Small-Footprint Keyword Spotting","display_name":"Disentangled Training with Adversarial Examples for Robust Small-Footprint Keyword Spotting","publication_year":2023,"publication_date":"2023-05-05","ids":{"openalex":"https://openalex.org/W4372191701","doi":"https://doi.org/10.1109/icassp49357.2023.10094750"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49357.2023.10094750","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp49357.2023.10094750","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5100344962","display_name":"Zhenyu Wang","orcid":"https://orcid.org/0000-0002-7291-4106"},"institutions":[{"id":"https://openalex.org/I162577319","display_name":"The University of Texas at Dallas","ror":"https://ror.org/049emcs32","country_code":"US","type":"education","lineage":["https://openalex.org/I162577319"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhenyu Wang","raw_affiliation_strings":["The University of Texas at Dallas,USA","The University of Texas at Dallas, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Texas at Dallas,USA","institution_ids":["https://openalex.org/I162577319"]},{"raw_affiliation_string":"The University of Texas at Dallas, USA","institution_ids":["https://openalex.org/I162577319"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029416441","display_name":"Li Wan","orcid":"https://orcid.org/0000-0002-1926-8758"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li Wan","raw_affiliation_strings":["META AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"META AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037276743","display_name":"Biqiao Zhang","orcid":"https://orcid.org/0000-0003-1598-2660"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Biqiao Zhang","raw_affiliation_strings":["META AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"META AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102111886","display_name":"Yiteng Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yiteng Huang","raw_affiliation_strings":["META AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"META AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029566548","display_name":"Shang-Wen Li","orcid":"https://orcid.org/0000-0003-0656-9874"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shang-Wen Li","raw_affiliation_strings":["META AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"META AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062411182","display_name":"Ming Sun","orcid":"https://orcid.org/0000-0001-7432-9704"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ming Sun","raw_affiliation_strings":["META AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"META AI","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101535850","display_name":"Xin Lei","orcid":"https://orcid.org/0000-0002-2466-3891"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xin Lei","raw_affiliation_strings":["META AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"META AI","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101951507","display_name":"Zhaojun Yang","orcid":"https://orcid.org/0000-0001-8707-3859"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhaojun Yang","raw_affiliation_strings":["META AI"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"META AI","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9995999932289124,"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":0.9995999932289124,"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/T11309","display_name":"Music and Audio Processing","score":0.9966999888420105,"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/T10028","display_name":"Topic Modeling","score":0.9952999949455261,"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":0.8300367593765259},{"id":"https://openalex.org/keywords/keyword-spotting","display_name":"Keyword spotting","score":0.8280274868011475},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.8149382472038269},{"id":"https://openalex.org/keywords/adversarial-system","display_name":"Adversarial system","score":0.7969610095024109},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5813117027282715},{"id":"https://openalex.org/keywords/spotting","display_name":"Spotting","score":0.5375108122825623},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.5052786469459534},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.4420994222164154},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4192262291908264},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3363696038722992}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8300367593765259},{"id":"https://openalex.org/C2781213101","wikidata":"https://www.wikidata.org/wiki/Q6398558","display_name":"Keyword spotting","level":2,"score":0.8280274868011475},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.8149382472038269},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.7969610095024109},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5813117027282715},{"id":"https://openalex.org/C2779506182","wikidata":"https://www.wikidata.org/wiki/Q7580141","display_name":"Spotting","level":2,"score":0.5375108122825623},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.5052786469459534},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.4420994222164154},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4192262291908264},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3363696038722992},{"id":"https://openalex.org/C127313418","wikidata":"https://www.wikidata.org/wiki/Q1069","display_name":"Geology","level":0,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C111368507","wikidata":"https://www.wikidata.org/wiki/Q43518","display_name":"Oceanography","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49357.2023.10094750","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp49357.2023.10094750","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":39,"referenced_works":["https://openalex.org/W114193738","https://openalex.org/W1549358575","https://openalex.org/W1945616565","https://openalex.org/W2034940213","https://openalex.org/W2056986588","https://openalex.org/W2079636521","https://openalex.org/W2101346879","https://openalex.org/W2122797512","https://openalex.org/W2407023693","https://openalex.org/W2748659049","https://openalex.org/W2752782242","https://openalex.org/W2797583228","https://openalex.org/W2797759721","https://openalex.org/W2807108439","https://openalex.org/W2884585870","https://openalex.org/W2936774411","https://openalex.org/W2938873567","https://openalex.org/W2963163009","https://openalex.org/W2963389226","https://openalex.org/W2963628261","https://openalex.org/W2964187693","https://openalex.org/W2964301649","https://openalex.org/W2971028215","https://openalex.org/W3007873534","https://openalex.org/W3025581723","https://openalex.org/W3035502324","https://openalex.org/W3035743198","https://openalex.org/W3166716987","https://openalex.org/W4293846201","https://openalex.org/W6632978964","https://openalex.org/W6640425456","https://openalex.org/W6729756640","https://openalex.org/W6739868092","https://openalex.org/W6750665317","https://openalex.org/W6751445234","https://openalex.org/W6753412334","https://openalex.org/W6763810570","https://openalex.org/W6774341790","https://openalex.org/W6796538260"],"related_works":["https://openalex.org/W2918559346","https://openalex.org/W2114097550","https://openalex.org/W4286904253","https://openalex.org/W3119978414","https://openalex.org/W2516975559","https://openalex.org/W2545741539","https://openalex.org/W3206647229","https://openalex.org/W1969408022","https://openalex.org/W2000885660","https://openalex.org/W2117995638"],"abstract_inverted_index":{"A":[0],"keyword":[1],"spotting":[2],"(KWS)":[3],"engine":[4],"continuously":[5],"running":[6],"on":[7,88,116,134],"the":[8,65,68,76,101,117,122,135],"device":[9],"is":[10,22,86],"exposed":[11],"to":[12,26,46,51,63,121],"various":[13],"speech":[14],"signals":[15],"that":[16,100],"are":[17],"usually":[18],"unseen":[19],"beforehand.":[20],"It":[21],"a":[23,28,93],"challenging":[24],"problem":[25],"build":[27],"small-footprint":[29],"and":[30,70,92],"high-performing":[31],"KWS":[32,53,83],"model":[33,84],"with":[34,60],"robustness":[35],"under":[36],"different":[37],"acoustic":[38],"environments.":[39],"In":[40],"this":[41],"paper,":[42],"we":[43],"explore":[44],"how":[45],"effectively":[47],"apply":[48],"adversarial":[49,61,71,126],"examples":[50,62],"improve":[52],"robustness.":[54],"We":[55],"propose":[56],"datasource-aware":[57],"disentangled":[58],"learning":[59,103],"reduce":[64],"mismatch":[66,77],"between":[67],"original":[69,79],"data":[72],"as":[73,75],"well":[74],"across":[78],"training":[80],"datasources.":[81],"The":[82],"architecture":[85],"based":[87],"depth-wise":[89],"separable":[90],"convolution":[91],"simple":[94],"attention":[95],"module.":[96],"Experimental":[97],"results":[98],"demonstrate":[99],"proposed":[102],"strategy":[104],"improves":[105],"false":[106,113],"reject":[107],"rate":[108,115],"by":[109],"40.31%":[110],"at":[111],"1%":[112],"accept":[114],"internal":[118],"dataset,":[119],"compared":[120],"strongest":[123],"baseline":[124],"without":[125],"examples.":[127],"Our":[128],"best-performing":[129],"system":[130],"achieves":[131],"98.06%":[132],"accuracy":[133],"Google":[136],"Speech":[137],"Commands":[138],"V1":[139],"dataset.":[140]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}