{"id":"https://openalex.org/W7225737800","doi":"https://doi.org/10.21203/rs.3.rs-11671474/v1","title":"Hiding a Voice From Speaker Encoders: Adversarial Protection Trained Over Measured Room Channels","display_name":"Hiding a Voice From Speaker Encoders: Adversarial Protection Trained Over Measured Room Channels","publication_year":2026,"publication_date":"2026-11-10","ids":{"openalex":"https://openalex.org/W7225737800","doi":"https://doi.org/10.21203/rs.3.rs-11671474/v1"},"language":"en","primary_location":{"id":"doi:10.21203/rs.3.rs-11671474/v1","doi":"https://doi.org/10.21203/rs.3.rs-11671474/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-11671474/v1","pdf_url":"https://www.researchsquare.com/article/rs-11671474/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","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":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://www.researchsquare.com/article/rs-11671474/latest.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5161362870","display_name":"Bhavishay Sheoran","orcid":"https://orcid.org/0009-0000-5298-5217","observed_orcids":["https://orcid.org/0009-0000-5298-5217"]},"institutions":[{"id":"https://openalex.org/I154851008","display_name":"Indian Institute of Technology Roorkee","ror":"https://ror.org/00582g326","country_code":"IN","type":"education","lineage":["https://openalex.org/I154851008"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Bhavishay Sheoran","raw_affiliation_strings":["Indian Institute of Technology Roorkee"],"raw_orcid":"https://orcid.org/0009-0000-5298-5217","affiliations":[{"raw_affiliation_string":"Indian Institute of Technology Roorkee","institution_ids":["https://openalex.org/I154851008"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5080620426","display_name":"Mayank Goswami","orcid":"https://orcid.org/0000-0001-8829-9980","observed_orcids":["https://orcid.org/0000-0001-8829-9980"]},"institutions":[{"id":"https://openalex.org/I154851008","display_name":"Indian Institute of Technology Roorkee","ror":"https://ror.org/00582g326","country_code":"IN","type":"education","lineage":["https://openalex.org/I154851008"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Mayank Goswami","raw_affiliation_strings":["Indian Institute of Technology Roorkee"],"raw_orcid":"https://orcid.org/0000-0001-8829-9980","affiliations":[{"raw_affiliation_string":"Indian Institute of Technology Roorkee","institution_ids":["https://openalex.org/I154851008"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I154851008"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.8314999938011169,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.8314999938011169,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.10760000348091125,"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":0.024900000542402267,"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":[],"concepts":[{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.661300003528595},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5825999975204468},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.5083000063896179},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.5042999982833862},{"id":"https://openalex.org/C103734657","wikidata":"https://www.wikidata.org/wiki/Q2739975","display_name":"PESQ","level":4,"score":0.44020000100135803},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.4221999943256378},{"id":"https://openalex.org/C169903167","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Test set","level":2,"score":0.39399999380111694},{"id":"https://openalex.org/C70836080","wikidata":"https://www.wikidata.org/wiki/Q837940","display_name":"Impulse (physics)","level":2,"score":0.38960000872612},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.3781000077724457},{"id":"https://openalex.org/C133892786","wikidata":"https://www.wikidata.org/wiki/Q1145189","display_name":"Speaker recognition","level":2,"score":0.3402000069618225},{"id":"https://openalex.org/C177291462","wikidata":"https://www.wikidata.org/wiki/Q423038","display_name":"Active listening","level":2,"score":0.3345000147819519},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3321000039577484},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.31690001487731934},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.3061999976634979},{"id":"https://openalex.org/C102894143","wikidata":"https://www.wikidata.org/wiki/Q1323979","display_name":"Monaural","level":2,"score":0.3000999987125397},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.2800999879837036},{"id":"https://openalex.org/C87619178","wikidata":"https://www.wikidata.org/wiki/Q126002","display_name":"Concatenation (mathematics)","level":2,"score":0.27619999647140503},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.26339998841285706}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21203/rs.3.rs-11671474/v1","doi":"https://doi.org/10.21203/rs.3.rs-11671474/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-11671474/v1","pdf_url":"https://www.researchsquare.com/article/rs-11671474/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","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":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"}],"best_oa_location":{"id":"doi:10.21203/rs.3.rs-11671474/v1","doi":"https://doi.org/10.21203/rs.3.rs-11671474/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-11671474/v1","pdf_url":"https://www.researchsquare.com/article/rs-11671474/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","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":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7225737800.pdf","grobid_xml":"https://content.openalex.org/works/W7225737800.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Abstract":[0],"Adversarial":[1],"perturbations":[2],"can":[3],"stop":[4],"a":[5,10,89,144,169,198,211,252,273],"speaker":[6],"encoder":[7,87],"from":[8,113,168],"extracting":[9],"talker\u2019s":[11],"identity.":[12],"The":[13],"tools":[14],"are":[15],"standard\u2014Projected":[16],"Gradient":[17],"Descent":[18],"(PGD),":[19],"Expectation":[20],"over":[21,72],"Transformation":[22],"(EOT),":[23],"and":[24,54,74,88,133,153,197,241,276],"room":[25,35],"impulse":[26],"responses\u2014but":[27],"the":[28,42,70,78,93,107,140,178,203,207,225,228,238,242,249,259,278,291],"closest":[29],"protection":[30,71,226],"methods":[31],"optimise":[32],"without":[33],"any":[34],"response,":[36],"so":[37],"it":[38,76,82,184,190],"is":[39],"open":[40],"whether":[41],"channel":[43,239],"model":[44],"used":[45],"in":[46,129,218,231,246,267,290],"optimisation":[47],"matters.":[48],"We":[49,57,80],"contribute":[50],"an":[51,84],"experimental":[52],"protocol":[53],"its":[55],"findings.":[56],"measured":[58,137,175,208],"four":[59,219],"acoustic":[60],"channels":[61,127,138],"by":[62,143,173,186],"exponential":[63],"sine":[64],"sweep":[65],"with":[66,214,256,286],"consumer":[67],"devices,":[68],"trained":[69,86],"three,":[73],"tested":[75],"on":[77,136,272],"fourth.":[79],"evaluated":[81],"using":[83],"independently":[85],"threshold":[90],"set":[91,201],"under":[92],"evaluation":[94],"conditions.":[95],"At":[96],"\u03b5":[97],"=":[98],"0.004":[99],"(SNR":[100],"23.9":[101],"dB,":[102],"PESQ":[103],"2.07,":[104],"ESTOI":[105],"0.86),":[106],"independent":[108],"encoder\u2019s":[109],"identification":[110],"rate":[111],"falls":[112],"98%":[114],"to":[115,270,284],"28%":[116],"(20":[117],"speakers":[118],"\u00d7":[119],"4":[120],"held-out":[121,179],"channels,":[122],"5":[123],"runs).":[124],"Against":[125],"simulated":[126,192,200],"matched":[128],"number,":[130],"reverberation":[131],"time,":[132],"post-processing,":[134],"training":[135],"lowers":[139,202],"judge\u2019s":[141,229],"similarity":[142,204,230],"further":[145,205],"0.045":[146],"(95%":[147],"CI":[148],"0.036\u20130.054)":[149],"at":[150,155],"equal":[151,156],"budget":[152],"0.039":[154],"SNR.":[157],"A":[158],"control":[159],"shows":[160],"that":[161],"about":[162],"80%":[163],"of":[164,183,222,248,293],"this":[165,223],"advantage":[166],"comes":[167],"spectral":[170],"colouring":[171],"shared":[172],"all":[174],"responses,":[176],"including":[177],"test":[180,213,255],"channel,":[181],"much":[182],"made":[185],"our":[187],"deconvolution.":[188],"With":[189],"added,":[191],"responses":[193],"come":[194],"within":[195],"0.007,":[196],"larger":[199],"than":[206,236],"ones.":[209],"In":[210,251],"feasibility":[212],"one":[215],"clip":[216],"re-recorded":[217],"rooms,":[220],"free":[221],"processing,":[224],"lowered":[227],"every":[232],"room,":[233],"but":[234],"less":[235,281],"through":[237],"models,":[240],"adversary":[243],"still":[244],"succeeded":[245],"three":[247],"four.":[250],"controlled":[253],"listening":[254],"28":[257],"listeners,":[258],"protected":[260],"speech":[261],"was":[262],"rated":[263],"1.2":[264],"points":[265],"lower":[266],"quality":[268],"(4.15":[269],"2.97":[271],"five-point":[274],"scale)":[275],"judged":[277],"same":[279],"talker":[280],"often":[282],"(76%":[283],"60%),":[285],"no":[287],"detectable":[288],"change":[289],"share":[292],"words":[294],"understood.":[295]},"counts_by_year":[],"updated_date":"2026-11-11T12:16:26.678958","created_date":"2026-11-11T00:00:00"}