{"id":"W2012902276","doi":"10.1109/icassp.2007.366951","title":"An Approach to Formant Frequency Estimation at Low Signal-to-Noise Ratio","year":2007,"lang":"en","type":"article","venue":"","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"","keywords":"Formant; Speech recognition; Computer science; TIMIT; Impulse response; White noise; Gaussian noise; Autoregressive model; Vocal tract; Noise (video); Signal-to-noise ratio (imaging); Speech processing; Acoustics; Algorithm; Mathematics; Artificial intelligence; Statistics; Hidden Markov model; Telecommunications; Physics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004222154,0.0001094866,0.00009297042,0.0001404204,0.0001752311,0.0002201984,0.0005180687,0.00003875614,0.00002321256],"category_scores_gemma":[0.0000184515,0.00009189956,0.00002380067,0.0004860683,0.000009049766,0.001017586,0.00009964469,0.00005081658,0.0002779816],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008351607,"about_ca_system_score_gemma":0.00004661153,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001815917,"about_ca_topic_score_gemma":0.00001511147,"domain_scores_codex":[0.9988626,0.0000115064,0.0002019005,0.000330182,0.0002691121,0.0003246614],"domain_scores_gemma":[0.9992437,0.0000191496,0.00004179651,0.0003475263,0.00007414294,0.000273663],"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":[0.00002075169,0.0002871879,0.001176268,0.00005050917,0.000008338656,0.00001534993,0.004365506,0.01607737,0.3595345,0.01117454,0.002187185,0.6051025],"study_design_scores_gemma":[0.0001760598,0.0001382304,0.003716803,0.00001743803,0.000002382536,0.00002083904,0.00005007593,0.2423451,0.7515131,0.001640411,0.0001108987,0.0002686455],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1103863,0.000009435054,0.8732228,0.0002157394,0.0000660178,0.0001779758,3.169616e-7,0.0001944344,0.01572696],"genre_scores_gemma":[0.506573,1.352758e-7,0.4921042,0.001026405,0.0000371671,0.000005848669,0.000003014408,0.000004030918,0.000246236],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6048339,"threshold_uncertainty_score":0.3747555,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0140445491084678,"score_gpt":0.2576774088132446,"score_spread":0.2436328597047768,"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."}}