{"id":"https://openalex.org/W4297841588","doi":"https://doi.org/10.21437/interspeech.2022-10895","title":"Extracting Targeted Training Data from ASR Models, and How to Mitigate It","display_name":"Extracting Targeted Training Data from ASR Models, and How to Mitigate It","publication_year":2022,"publication_date":"2022-09-16","ids":{"openalex":"https://openalex.org/W4297841588","doi":"https://doi.org/10.21437/interspeech.2022-10895"},"language":"en","primary_location":{"id":"doi:10.21437/interspeech.2022-10895","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10895","pdf_url":null,"source":{"id":"https://openalex.org/S4363604309","display_name":"Interspeech 2022","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":"Interspeech 2022","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/A5056776503","display_name":"Ehsan Amid","orcid":"https://orcid.org/0000-0001-6097-0226"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Ehsan Amid","raw_affiliation_strings":["Google LLC"],"affiliations":[{"raw_affiliation_string":"Google LLC","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077162569","display_name":"Om Thakkar","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Om Dipakbhai Thakkar","raw_affiliation_strings":["Google LLC"],"affiliations":[{"raw_affiliation_string":"Google LLC","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000078382","display_name":"Arun Narayanan","orcid":"https://orcid.org/0009-0008-3325-8928"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Arun Narayanan","raw_affiliation_strings":["Google LLC"],"affiliations":[{"raw_affiliation_string":"Google LLC","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075324202","display_name":"Rajiv Mathews","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rajiv Mathews","raw_affiliation_strings":["Google LLC"],"affiliations":[{"raw_affiliation_string":"Google LLC","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110711540","display_name":"Fran\u00e7oise Beaufays","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Francoise Beaufays","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5056776503"],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":0.3122,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.50864623,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"2803","last_page":"2807"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.8101999759674072,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.8101999759674072,"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/training","display_name":"Training (meteorology)","score":0.748848557472229},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7422385811805725},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.5629291534423828},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.48112645745277405},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.36535894870758057},{"id":"https://openalex.org/keywords/database","display_name":"Database","score":0.14284616708755493}],"concepts":[{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.748848557472229},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7422385811805725},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.5629291534423828},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.48112645745277405},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36535894870758057},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.14284616708755493},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21437/interspeech.2022-10895","is_oa":false,"landing_page_url":"https://doi.org/10.21437/interspeech.2022-10895","pdf_url":null,"source":{"id":"https://openalex.org/S4363604309","display_name":"Interspeech 2022","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":"Interspeech 2022","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W1494198834","https://openalex.org/W1992926795","https://openalex.org/W2027595342","https://openalex.org/W2473418344","https://openalex.org/W2617258110","https://openalex.org/W2946930197","https://openalex.org/W2952604841","https://openalex.org/W2963122170","https://openalex.org/W3015995734","https://openalex.org/W3035034338","https://openalex.org/W3048045781","https://openalex.org/W3097777922","https://openalex.org/W3172323480","https://openalex.org/W3197956343","https://openalex.org/W3210259840","https://openalex.org/W3211753216","https://openalex.org/W4221159672","https://openalex.org/W4287553002","https://openalex.org/W4287663285","https://openalex.org/W4294619240","https://openalex.org/W4303580678"],"related_works":["https://openalex.org/W2748952813","https://openalex.org/W230091440","https://openalex.org/W2390279801","https://openalex.org/W2233261550","https://openalex.org/W2358668433","https://openalex.org/W2551249631","https://openalex.org/W3158596343","https://openalex.org/W4285322112","https://openalex.org/W4292794239","https://openalex.org/W4385572030"],"abstract_inverted_index":{"Recent":[0],"work":[1],"has":[2],"designed":[3],"methods":[4],"to":[5,33,95],"demonstrate":[6,34,62],"that":[7,91,123,128,159],"model":[8,109],"updates":[9],"in":[10,22,71,125,164],"ASR":[11,42,60],"training":[12,38,56,83,102,165],"can":[13],"leak":[14],"potentially":[15],"sensitive":[16],"attributes":[17],"of":[18,55,65,118],"the":[19,24,30,63,78,97,108,114,119,136,175,186],"utterances":[20,103],"used":[21,81,163],"computing":[23],"updates.In":[25],"this":[26],"work,":[27],"we":[28,89,92,121,151,160],"design":[29,44,152],"first":[31],"method":[32,50,140,158],"information":[35],"leakage":[36],"about":[37],"data":[39,57,156],"from":[40,58,77,100,113,135],"trained":[41,59],"models.We":[43,61],"Noise":[45,66,183],"Masking,":[46],"a":[47,84,126,155],"fill-in-the-blank":[48],"style":[49],"for":[51,74,82],"extracting":[52,75],"targeted":[53],"parts":[54],"success":[64,149],"Masking":[67,184],"by":[68],"using":[69],"it":[70],"four":[72,187],"settings":[73],"names":[76,99],"LibriSpeech":[79],"dataset":[80],"state-of-the-art":[85],"Conformer":[86],"model.In":[87],"particular,":[88],"show":[90,122,161],"are":[93],"able":[94],"extract":[96],"correct":[98,143],"masked":[101],"with":[104,167,178],"11.8%":[105],"accuracy,":[106],"while":[107],"outputs":[110],"some":[111],"name":[112,144,148],"train":[115],"set":[116],"55.2%":[117],"time.Further,":[120],"even":[124],"setting":[127],"uses":[129],"synthetic":[130],"audio":[131],"and":[132],"partial":[133],"transcripts":[134],"test":[137],"set,":[138],"our":[139],"achieves":[141],"2.5%":[142],"accuracy":[145],"(47.7%":[146],"any":[147],"rate).Lastly,":[150],"Word":[153],"Dropout,":[154],"augmentation":[157],"when":[162],"along":[166,177],"Multistyle":[168],"TRaining":[169],"(MTR),":[170],"provides":[171],"comparable":[172],"utility":[173],"as":[174],"baseline,":[176],"significantly":[179],"mitigating":[180],"extraction":[181],"via":[182],"across":[185],"evaluated":[188],"settings.":[189]},"counts_by_year":[{"year":2024,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}