{"id":"W6921056807","doi":"10.60864/1s30-t715","title":"Less peaky and more accurate CTC forced alignment by label priors","year":2024,"lang":"en","type":"article","venue":"IEEE SIGPORT","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Offset (computer science); Security token; Pipeline (software); Hidden Markov model; Pattern recognition (psychology); Bayesian probability; Prior probability; Training set","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002191119,0.0001365856,0.000136536,0.00007406959,0.00007235015,0.0002807436,0.0002813793,0.00005747842,0.00004393894],"category_scores_gemma":[0.00001322807,0.0001171361,0.00004474432,0.0001947209,0.00004099868,0.0003169667,0.00006616156,0.00008743506,0.00009520296],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003149046,"about_ca_system_score_gemma":0.00005664313,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002537268,"about_ca_topic_score_gemma":0.000004790911,"domain_scores_codex":[0.9989043,0.00002338648,0.0001873777,0.0003980007,0.0002641124,0.0002228064],"domain_scores_gemma":[0.9994785,0.00007337996,0.00003505941,0.0002606198,0.00003116179,0.0001212979],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00000668588,0.00008469633,0.000259015,0.00008667497,0.00009881231,0.0003452815,0.001586031,0.000003572944,0.02967662,0.003480563,0.08644681,0.8779252],"study_design_scores_gemma":[0.001702568,0.0003267333,0.002604217,0.0004963196,0.0001322367,0.0005672093,0.001651372,0.1879608,0.5106731,0.004821382,0.2869973,0.002066762],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8200011,0.00110561,0.1591778,0.005059347,0.002157193,0.0004623052,0.00004222888,0.0008354922,0.01115896],"genre_scores_gemma":[0.98722,0.0001514617,0.006006412,0.001008686,0.00009354307,0.00003997947,0.000007085172,0.00002042633,0.005452428],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8758585,"threshold_uncertainty_score":0.4776673,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03190628355525705,"score_gpt":0.281614336227028,"score_spread":0.2497080526717709,"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."}}