{"id":"https://openalex.org/W7148264062","doi":"https://doi.org/10.1109/asru65441.2025.11434609","title":"DarkStream: real-time speech anonymization with low latency","display_name":"DarkStream: real-time speech anonymization with low latency","publication_year":2025,"publication_date":"2025-12-06","ids":{"openalex":"https://openalex.org/W7148264062","doi":"https://doi.org/10.1109/asru65441.2025.11434609"},"language":null,"primary_location":{"id":"doi:10.1109/asru65441.2025.11434609","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru65441.2025.11434609","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","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/A5043014650","display_name":"Waris Quamer","orcid":"https://orcid.org/0000-0001-9087-0242"},"institutions":[{"id":"https://openalex.org/I91045830","display_name":"Texas A&M University","ror":"https://ror.org/01f5ytq51","country_code":"US","type":"education","lineage":["https://openalex.org/I91045830"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Waris Quamer","raw_affiliation_strings":["Texas A&M University,Dept. of Computer Science and Engineering,College Station,US"],"affiliations":[{"raw_affiliation_string":"Texas A&M University,Dept. of Computer Science and Engineering,College Station,US","institution_ids":["https://openalex.org/I91045830"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5129665034","display_name":"Ricardo Gutierrez-Osuna","orcid":null},"institutions":[{"id":"https://openalex.org/I91045830","display_name":"Texas A&M University","ror":"https://ror.org/01f5ytq51","country_code":"US","type":"education","lineage":["https://openalex.org/I91045830"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ricardo Gutierrez-Osuna","raw_affiliation_strings":["Texas A&M University,Dept. of Computer Science and Engineering,College Station,US"],"affiliations":[{"raw_affiliation_string":"Texas A&M University,Dept. of Computer Science and Engineering,College Station,US","institution_ids":["https://openalex.org/I91045830"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5043014650"],"corresponding_institution_ids":["https://openalex.org/I91045830"],"apc_list":null,"apc_paid":null,"fwci":2.1819,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.92935426,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10201","display_name":"Speech Recognition and Synthesis","score":0.4602999985218048,"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":0.4602999985218048,"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/T11598","display_name":"Internet Traffic Analysis and Secure E-voting","score":0.19460000097751617,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.08460000157356262,"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/low-latency","display_name":"Low latency (capital markets)","score":0.2973000109195709},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.2524999976158142},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.2522999942302704},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.24799999594688416},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.24420000612735748}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6707000136375427},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3855000138282776},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3292999863624573},{"id":"https://openalex.org/C46637626","wikidata":"https://www.wikidata.org/wiki/Q6693015","display_name":"Low latency (capital markets)","level":2,"score":0.2973000109195709},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.2524999976158142},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2522999942302704},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.24799999594688416},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.24420000612735748},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.23770000040531158},{"id":"https://openalex.org/C104267543","wikidata":"https://www.wikidata.org/wiki/Q208163","display_name":"Signal processing","level":3,"score":0.23659999668598175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asru65441.2025.11434609","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asru65441.2025.11434609","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320333051","display_name":"Intelligence Advanced Research Projects Activity","ror":"https://ror.org/01v3fsc55"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":17,"referenced_works":["https://openalex.org/W2146334809","https://openalex.org/W2603777577","https://openalex.org/W2760103357","https://openalex.org/W2890964092","https://openalex.org/W3024768724","https://openalex.org/W3024869864","https://openalex.org/W3043999252","https://openalex.org/W3197008493","https://openalex.org/W4206231964","https://openalex.org/W4221114270","https://openalex.org/W4292597003","https://openalex.org/W4297841819","https://openalex.org/W4319862433","https://openalex.org/W4386557792","https://openalex.org/W4399485371","https://openalex.org/W4406461813","https://openalex.org/W7133219565"],"related_works":[],"abstract_inverted_index":{"We":[0],"propose":[1],"DarkStream,":[2],"a":[3,22,26,45,60,111],"streaming":[4],"speech":[5,117],"synthesis":[6],"model":[7,40,74],"for":[8,114],"real-time":[9,116],"speaker":[10,56,82],"anonymization.":[11],"To":[12,34],"improve":[13],"content":[14,69],"encoding":[15],"under":[16],"strict":[17],"latency":[18],"constraints,":[19],"DarkStream":[20,54,109],"combines":[21],"causal":[23],"waveform":[24],"encoder,":[25],"short":[27],"lookahead":[28],"buffer,":[29],"and":[30,105],"transformer-based":[31],"contextual":[32],"layers.":[33],"further":[35],"reduce":[36],"inference":[37],"time,":[38],"the":[39,68,88],"generates":[41],"waveforms":[42],"directly":[43],"via":[44],"neural":[46],"vocoder,":[47],"thus":[48],"removing":[49],"intermediate":[50],"mel-spectrogram":[51],"conversions.":[52],"Finally,":[53],"anonymizes":[55],"identity":[57],"by":[58],"injecting":[59],"GAN-generated":[61],"pseudo-speaker":[62],"embedding":[63],"into":[64],"linguistic":[65,95],"features":[66],"from":[67],"encoder.":[70],"Evaluations":[71],"show":[72],"our":[73],"achieves":[75],"strong":[76],"anonymization,":[77],"yielding":[78],"close":[79],"to":[80],"50%":[81],"verification":[83],"EER":[84],"(near-chance":[85],"performance)":[86],"on":[87],"lazy-informed":[89],"attack":[90],"scenario,":[91],"while":[92],"maintaining":[93],"acceptable":[94],"intelligibility":[96,107],"(WER":[97],"within":[98],"9%).":[99],"By":[100],"balancing":[101],"low-latency,":[102],"robust":[103],"privacy,":[104],"minimal":[106],"degradation,":[108],"provides":[110],"practical":[112],"solution":[113],"privacy-preserving":[115],"communication.":[118]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-04-23T09:07:50.710637","created_date":"2026-04-03T00:00:00"}