{"id":"W4386566564","doi":"10.18653/v1/2023.fieldmatters-1.4","title":"Speech Database (Speech-DB) – An on-line platform for storing, validating, searching, and recording spoken language data","year":2023,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Social Sciences and Humanities Research Council of Canada; University of Alberta","keywords":"Computer science; Spoken language; Documentation; Speech corpus; Speech recognition; The Internet; Line (geometry); Speech synthesis; Database; Speech processing; Natural language processing; World Wide Web; Programming language","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003081152,0.001545016,0.001154526,0.002084021,0.0008388099,0.003034539,0.002057913,0.001021521,0.06477644],"category_scores_gemma":[0.008670549,0.0008023121,0.0007423182,0.001356235,0.0007186459,0.003959057,0.00419458,0.001297152,0.08586894],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007840093,"about_ca_system_score_gemma":0.002402246,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008531438,"about_ca_topic_score_gemma":0.008450896,"domain_scores_codex":[0.9977964,0.0003656773,0.0002817105,0.0004959551,0.000906761,0.0001535445],"domain_scores_gemma":[0.993946,0.001846489,0.0003475517,0.00142409,0.001824573,0.000611304],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.002184413,0.0002051586,0.002830167,0.001835994,0.0002020991,0.0007898401,0.002362706,0.0011269,0.03551299,0.01109701,0.6696077,0.272245],"study_design_scores_gemma":[0.000374085,0.0002519655,0.007582295,0.0003680341,0.0001123207,0.00153062,0.0008306958,0.01567109,0.06446662,0.009864809,0.8985597,0.0003878016],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"software","genre_gemma":"software","genre_scores_codex":[0.0151073,0.001292501,0.2512473,0.0007361195,0.0006762399,0.001389346,0.1479723,0.517274,0.06430489],"genre_scores_gemma":[0.1187933,0.001613463,0.2983981,0.001440762,0.0005072781,0.002534106,0.4175717,0.07263842,0.08650283],"genre_candidate":"software","genre_consensus":"software","teacher_disagreement_score":0.06477644,"threshold_uncertainty_score":0.2166988,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1241002357902389,"score_gpt":0.3997001292812089,"score_spread":0.27559989349097,"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."}}