{"id":"W34011602","doi":"10.21437/interspeech.2006-40","title":"Measuring the acceptable word error rate of machine-generated webcast transcripts","year":2006,"lang":"en","type":"article","venue":"","topic":"Speech and dialogue systems","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University; University of Toronto","funders":"","keywords":"Webcast; Computer science; Broadcasting (networking); Word (group theory); Metadata; The Internet; Speech recognition; Information retrieval; World Wide Web; Linguistics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"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":[0.008477107,0.0009221498,0.0008144427,0.001659039,0.0005674323,0.001364175,0.0008149776,0.001495269,0.001519399],"category_scores_gemma":[0.08615994,0.0003290758,0.000560226,0.001148967,0.0007509352,0.001292758,0.001024586,0.0007972186,0.001485945],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004178026,"about_ca_system_score_gemma":0.0004273684,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009092639,"about_ca_topic_score_gemma":0.0008332069,"domain_scores_codex":[0.9871321,0.005428633,0.002331304,0.001978664,0.002756991,0.0003723779],"domain_scores_gemma":[0.9103997,0.06625105,0.003653532,0.005240025,0.01394986,0.0005058265],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.004971741,0.0009081688,0.06506085,0.00161827,0.0004221158,0.00143592,0.01061699,0.03136069,0.5340838,0.002267197,0.00273705,0.3445172],"study_design_scores_gemma":[0.0002693261,0.005530937,0.1415629,0.0001494636,0.0003797082,0.002805922,0.003418795,0.2058646,0.6303059,0.003326376,0.005956043,0.0004301176],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8031977,0.0002477928,0.1900976,0.0001566261,0.0001425815,0.0005341968,0.001182196,0.002320379,0.002120884],"genre_scores_gemma":[0.8615445,0.0001714633,0.1335983,0.00007651999,0.00005388332,0.0009124776,0.0020097,0.0005112693,0.001121837],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.008477107,"threshold_uncertainty_score":0.04483175,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03675406798063769,"score_gpt":0.2255474388374474,"score_spread":0.1887933708568097,"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."}}