{"id":"W4367145625","doi":"10.1121/10.0019208","title":"Using the Corpus of North American Spoken English to explore regional variation in millions of eɪ diphthongs","year":2023,"lang":"en","type":"article","venue":"The Journal of the Acoustical Society of America","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; American English; Python (programming language); Diphthong; Speech recognition; Speech corpus; Coarticulation; Natural language processing; Variation (astronomy); Variety (cybernetics); Artificial intelligence; Speech synthesis; Linguistics; Vowel","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008256189,0.0004554175,0.0003599418,0.002864175,0.001483076,0.001041559,0.0005803646,0.0003691748,0.009693717],"category_scores_gemma":[0.003842538,0.0002482389,0.0002864764,0.003584563,0.0006750797,0.0007996582,0.001525215,0.0006463536,0.004587185],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008991097,"about_ca_system_score_gemma":0.00201592,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.05985274,"about_ca_topic_score_gemma":0.1468068,"domain_scores_codex":[0.9989113,0.0002213695,0.0001206784,0.0003490476,0.0003056152,0.00009212709],"domain_scores_gemma":[0.9967038,0.00122136,0.0001827263,0.0003154176,0.001417655,0.000159102],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0008601849,0.0003967919,0.04885095,0.004945698,0.0002051804,0.003466028,0.03915242,0.002227067,0.09769169,0.007195247,0.2739007,0.521108],"study_design_scores_gemma":[0.00006620983,0.0001276228,0.4447515,0.0003762177,0.0001478412,0.001623143,0.01668017,0.003959698,0.0131746,0.001479289,0.5174512,0.0001625059],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6205829,0.00477954,0.03870444,0.001524607,0.0008525599,0.001228306,0.2552734,0.002677182,0.07437702],"genre_scores_gemma":[0.5664545,0.002429429,0.06498893,0.0006358463,0.0003766859,0.003787123,0.3311854,0.001569853,0.0285723],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.05985274,"threshold_uncertainty_score":0.1190087,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06297583373034152,"score_gpt":0.2886126725932716,"score_spread":0.2256368388629301,"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."}}