{"id":"W3217533145","doi":"10.1121/10.0008580","title":"Exploring the variable efficacy of Google speech-to-text with spontaneous bilingual speech in Cantonese and English","year":2021,"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":"University of British Columbia","funders":"","keywords":"Computer science; Context (archaeology); Variation (astronomy); Variable (mathematics); Matching (statistics); Speech recognition; Multilingualism; Natural language processing; Linguistics; Psychology; Mathematics; History; Statistics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001023981,0.0001308855,0.0003612463,0.00002945889,0.0001228664,0.00004427142,0.0009107806,0.00003484027,0.00004638423],"category_scores_gemma":[0.001172344,0.00006175693,0.0001604114,0.000882472,0.0003469435,0.0001436902,0.0003429426,0.0003982803,0.000001457475],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005246591,"about_ca_system_score_gemma":0.0003026534,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001530089,"about_ca_topic_score_gemma":0.00001322843,"domain_scores_codex":[0.9983392,0.0002619164,0.0004311435,0.0001458037,0.0005777304,0.0002442355],"domain_scores_gemma":[0.996608,0.002134693,0.0002958267,0.0004568233,0.0004054724,0.00009920921],"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":[0.0007941204,0.001278278,0.0004997727,0.0001478802,0.00062745,0.0004443144,0.03411742,0.01142247,0.06129222,0.0002198876,0.002456614,0.8866996],"study_design_scores_gemma":[0.01049446,0.004978788,0.0448708,0.004316465,0.002187246,0.03356371,0.1272256,0.4221435,0.3217142,0.006764951,0.01924325,0.002497031],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9100484,0.00018162,0.08446567,0.00427764,0.0003047124,0.0001812478,0.000005446989,0.00001355101,0.0005217284],"genre_scores_gemma":[0.7120849,0.0004114625,0.2862418,0.0009970407,0.0001525672,0.000001387078,1.112206e-7,0.00001340311,0.00009726785],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8842025,"threshold_uncertainty_score":0.2518374,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02635811481932245,"score_gpt":0.2388873184202648,"score_spread":0.2125292036009424,"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."}}