{"id":"W3029693316","doi":"","title":"Speech Transcription Challenges for Resource Constrained Indigenous Language Cree","year":2020,"lang":"en","type":"article","venue":"Workshop Spoken Language Technologies for Under-resourced Languages","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Word error rate; Transcription (linguistics); Speech recognition; Natural language processing; Word (group theory); Language model; Documentation; Artificial intelligence; Spoken language; Error analysis; Linguistics","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":[0.00237153,0.0005009483,0.0008586257,0.0005612224,0.001489113,0.002192854,0.001295728,0.001731657,0.004600545],"category_scores_gemma":[0.01267927,0.0003309932,0.0004063922,0.000808708,0.000971753,0.001482107,0.001427921,0.001481483,0.003463483],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001440381,"about_ca_system_score_gemma":0.003199255,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.1059594,"about_ca_topic_score_gemma":0.1541652,"domain_scores_codex":[0.9975554,0.0006656584,0.0002076995,0.0005820958,0.0007528883,0.0002361723],"domain_scores_gemma":[0.991675,0.00418402,0.0001706663,0.0008664613,0.002839275,0.000264549],"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":[0.001364146,0.0002971025,0.0106938,0.001927273,0.0002247432,0.00451467,0.02138108,0.04758326,0.2463362,0.004536004,0.03536923,0.6257726],"study_design_scores_gemma":[0.0004518272,0.0008256621,0.04976805,0.0008033625,0.0004323281,0.009776888,0.06214437,0.3343771,0.2760676,0.01220833,0.2523065,0.0008381027],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7275562,0.003228255,0.2189673,0.009161358,0.0007441631,0.000395151,0.006477325,0.01163723,0.0218329],"genre_scores_gemma":[0.8831086,0.001549734,0.0884839,0.0015404,0.0002558575,0.0003149798,0.008326894,0.001217337,0.01520246],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1059594,"threshold_uncertainty_score":0.2106852,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04157063800353791,"score_gpt":0.2814875973299665,"score_spread":0.2399169593264286,"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."}}