{"id":"W4281250109","doi":"10.56553/popets-2023-0007","title":"Differentially Private Speaker Anonymization","year":2023,"lang":"en","type":"article","venue":"Proceedings on Privacy Enhancing Technologies","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Toronto","funders":"Defense Advanced Research Projects Agency; Canadian Institute for Advanced Research; Centre National de la Recherche Scientifique; Alan Turing Institute; European Commission; Agence Nationale de la Recherche","keywords":"Computer science; Speaker diarisation; Speech recognition; Utterance; Speaker recognition; Embedding; Inference; Natural language processing; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","metaepi_narrow","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.0007373736,0.0005171245,0.0004508487,0.001408253,0.0003950369,0.0006371991,0.03551101,0.0004904304,0.00001026715],"category_scores_gemma":[0.04796918,0.0004744609,0.0001142408,0.00411014,0.0002313331,0.001597016,0.09586057,0.0007719637,0.0006178432],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002062023,"about_ca_system_score_gemma":0.00005246283,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003857053,"about_ca_topic_score_gemma":0.00000177679,"domain_scores_codex":[0.9959829,0.00001495245,0.0006239404,0.001451034,0.0008058745,0.001121322],"domain_scores_gemma":[0.9935665,0.0001916234,0.0003825368,0.00557571,0.0002176982,0.00006594403],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00004655063,0.0002648684,0.006291906,0.0003511941,0.0001438403,0.00007550023,0.0005804644,0.0000215751,0.2011518,0.2483328,0.2734929,0.2692466],"study_design_scores_gemma":[0.0003239905,0.0001807093,0.001746244,0.0002029077,0.000009354706,0.00001427006,0.0001875448,0.02646872,0.4521884,0.5120994,0.005999128,0.0005793259],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6308701,0.0001008537,0.2238724,0.06110227,0.0008386514,0.0008630786,0.00001056212,0.08038201,0.001960035],"genre_scores_gemma":[0.8059412,0.0004410158,0.1930724,0.0001360928,0.00004336284,0.0001687033,0.0000143369,0.00005819301,0.000124657],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2686673,"threshold_uncertainty_score":0.9997707,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02373379509920797,"score_gpt":0.2614776336509623,"score_spread":0.2377438385517543,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}