{"id":"W4408352315","doi":"10.1109/icassp49660.2025.10888434","title":"Large-Scale Recurrent Neural Networks with Fully Homomorphic Encryption for Privacy-Enhanced Speaker Identification","year":2025,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Homomorphic encryption; Computer science; Encryption; Identification (biology); Speaker identification; Artificial neural network; Scale (ratio); Speech recognition; Computer security; Computer network; Speaker recognition; 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":[],"consensus_categories":[],"category_scores_codex":[0.000324891,0.0001317994,0.000145474,0.0001276614,0.000183649,0.0002497124,0.000444707,0.00006395468,0.00009394178],"category_scores_gemma":[0.00004815884,0.000107197,0.00007647624,0.0004867636,0.00002478266,0.0003894422,0.00008229305,0.00009326746,0.00005257426],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000421913,"about_ca_system_score_gemma":0.00003762699,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002339929,"about_ca_topic_score_gemma":0.00005229868,"domain_scores_codex":[0.998803,0.0000524325,0.0002664468,0.0004354707,0.0001717287,0.0002709428],"domain_scores_gemma":[0.9990675,0.00009609653,0.00009958456,0.0004572368,0.0002207081,0.00005886555],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001566473,0.0005109292,0.00038812,0.00007745949,0.00007777212,0.000003553379,0.000754956,0.0004163074,0.004164228,0.02442073,0.01463106,0.9543982],"study_design_scores_gemma":[0.0009496192,0.000110128,0.006627573,0.00006575672,0.00003128988,0.000006602511,0.0001408515,0.9640527,0.01872401,0.002698954,0.00631306,0.0002794034],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0221923,0.00004327247,0.9724599,0.001638832,0.0006187444,0.0005092133,0.000002912238,0.0003360503,0.002198816],"genre_scores_gemma":[0.9025742,0.00003014801,0.09329775,0.0007168362,0.00008833739,0.000234775,0.00002420012,0.0000134056,0.003020379],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9636365,"threshold_uncertainty_score":0.4371367,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01454853176168504,"score_gpt":0.2527619497386806,"score_spread":0.2382134179769956,"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."}}