{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008458487,0.0001000995,0.0003273684,0.0000658295,0.0001074857,0.0000132667,0.001170483,0.00002503468,0.000007276832],"category_scores_gemma":[0.0006225885,0.00005203708,0.000317334,0.001994982,0.0005162263,0.0001068989,0.0002682054,0.000251986,0.000001465452],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005001816,"about_ca_system_score_gemma":0.0001157713,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001984184,"about_ca_topic_score_gemma":0.00000604627,"domain_scores_codex":[0.9982501,0.0002720466,0.0005346049,0.00009882018,0.000651571,0.0001928329],"domain_scores_gemma":[0.9971204,0.001295145,0.000760383,0.0003624819,0.0003945541,0.00006697728],"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":[0.0007794948,0.001600739,0.01671898,0.000202576,0.001251659,0.00001509321,0.142387,0.1529465,0.09975946,0.0003332081,0.06096805,0.5230373],"study_design_scores_gemma":[0.00063041,0.0005761011,0.3413571,0.0002986318,0.0002802903,0.00005991281,0.01425642,0.6351541,0.004125635,0.00165142,0.001327383,0.0002826357],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5957384,0.00001989822,0.3971893,0.006647281,0.0002086343,0.0001435646,0.00001004042,0.00001231877,0.00003058267],"genre_scores_gemma":[0.8945886,0.0002521204,0.1040027,0.001016648,0.0001163844,0.000001649528,2.80766e-7,0.000008904217,0.0000127104],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5227547,"threshold_uncertainty_score":0.2175068,"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."}}