{"id":"W4386896658","doi":"10.1038/s41598-023-42818-3","title":"Comparison of the prediction accuracy of machine learning algorithms in crosslinguistic vowel classification","year":2023,"lang":"en","type":"article","venue":"Scientific Reports","topic":"Speech and Audio Processing","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Humanities Research Group, University of Windsor; University of Nicosia","keywords":"Formant; Vowel; Linear discriminant analysis; Computer science; Speech recognition; Artificial neural network; Artificial intelligence; Perception; Decision tree; Similarity (geometry); Natural language processing; Psychology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001514748,0.00007164966,0.0001527279,0.0002208175,0.0002344694,0.0001506369,0.0003772885,0.00003950536,0.000004184383],"category_scores_gemma":[0.001595299,0.00005481362,0.00005415382,0.002123124,0.0001624759,0.000218727,0.0001784142,0.0001505318,0.000004093954],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002960572,"about_ca_system_score_gemma":0.0001555261,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003740435,"about_ca_topic_score_gemma":0.00001961489,"domain_scores_codex":[0.9981779,0.00005522354,0.0006388593,0.0004084181,0.0005380524,0.0001815443],"domain_scores_gemma":[0.9983293,0.0000869284,0.0007400543,0.0006095351,0.0002031269,0.00003100944],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000003740787,0.00013632,0.7100587,0.0001140359,0.000007549826,0.00002808205,0.005286265,0.008114295,0.2042332,0.0001171055,0.001117846,0.07078288],"study_design_scores_gemma":[0.0001046951,0.0000200501,0.2242802,0.0001345918,0.000004916034,0.00001994795,0.0001478083,0.4118385,0.3575045,0.004134415,0.001730126,0.00008025396],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9709255,0.000141254,0.02206215,0.0001913615,0.005794338,0.0002090389,0.000001794945,0.0001372101,0.0005374103],"genre_scores_gemma":[0.9972395,0.000002276411,0.002216349,0.000003063016,0.00002407563,0.000005741896,0.00001270743,0.000004456899,0.0004918202],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4857785,"threshold_uncertainty_score":0.2235235,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04664852449512247,"score_gpt":0.3345721311493576,"score_spread":0.2879236066542352,"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."}}