{"id":"W2914422976","doi":"10.48550/arxiv.1902.02375","title":"Centroid-based deep metric learning for speaker recognition","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Embedding; Computer science; Centroid; Speech recognition; Similarity (geometry); Task (project management); Artificial intelligence; Metric (unit); Set (abstract data type); Speaker recognition; Pattern recognition (psychology); Speaker diarisation; Natural language processing; Image (mathematics)","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001147728,0.0009384025,0.001061409,0.0006848404,0.0003490174,0.0006735689,0.001663056,0.0009238489,0.003904925],"category_scores_gemma":[0.002851805,0.0003773898,0.0006177336,0.0007312825,0.0005977007,0.001752513,0.001711841,0.001985772,0.002212038],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001142673,"about_ca_system_score_gemma":0.0008040553,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006314558,"about_ca_topic_score_gemma":0.007340162,"domain_scores_codex":[0.9992656,0.0002407273,0.00003489301,0.0002078073,0.000182629,0.00006829179],"domain_scores_gemma":[0.9994784,0.0001573369,0.00005056123,0.0001137476,0.0001582604,0.00004179498],"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":[0.0002827939,0.000126175,0.0008112091,0.0001402608,0.0001380301,0.00006446994,0.0001487564,0.2649578,0.0157725,0.03042712,0.01340262,0.6737283],"study_design_scores_gemma":[0.000006197526,0.00003655137,0.0002142222,0.000006452845,0.000009955483,0.00003189159,0.00001211127,0.9764134,0.003498919,0.01803227,0.001726057,0.00001190596],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01109555,0.001447232,0.9832169,0.0002458995,0.00009734964,0.00002475669,0.0001955707,0.002334546,0.001342102],"genre_scores_gemma":[0.592622,0.001332442,0.3890028,0.0003195125,0.0002588897,0.0001498309,0.002043417,0.0006890015,0.01358213],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006314558,"threshold_uncertainty_score":0.01306331,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09792118983636798,"score_gpt":0.1929767119206509,"score_spread":0.09505552208428293,"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."}}