{"id":"W3015280134","doi":"10.1109/icassp40776.2020.9053831","title":"Improving Speech Recognition Using Consistent Predictions on Synthesized Speech","year":2020,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":51,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Speech recognition; Computer science; Audio mining; Voice activity detection; Acoustic model; Speech processing; Point (geometry); Speech synthesis; Speech coding; Natural language processing; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0002511243,0.0002004895,0.0002335663,0.0001456449,0.0002258608,0.0002513791,0.0003956641,0.00009877252,0.0006047774],"category_scores_gemma":[0.0005538937,0.0001852337,0.0001600954,0.000480647,0.00004117992,0.000468424,0.0001189417,0.0001882127,0.0009108662],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008002658,"about_ca_system_score_gemma":0.00009422079,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000770504,"about_ca_topic_score_gemma":0.000009142824,"domain_scores_codex":[0.9982377,0.0001256021,0.0003527345,0.0005779664,0.0003998344,0.0003061864],"domain_scores_gemma":[0.998849,0.0002943599,0.0001263853,0.0003504707,0.0001121034,0.0002676606],"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":[0.00002912206,0.0001121808,0.00004139594,0.00001839301,0.00003948229,0.00007755263,0.0001139637,0.00000770737,0.01497868,0.0003665877,0.00104043,0.9831745],"study_design_scores_gemma":[0.00076721,0.0002338222,0.000106785,0.00008615061,0.00005923564,0.0002156525,0.0002351494,0.6745013,0.3204057,0.0009098196,0.001963939,0.0005152714],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08928914,0.00002549133,0.8499635,0.007505929,0.0007835526,0.0006841648,0.00003028453,0.001676089,0.05004187],"genre_scores_gemma":[0.2877926,0.00001684792,0.7053628,0.006275066,0.0003137233,0.00002295688,0.000008741444,0.00002994009,0.000177363],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9826592,"threshold_uncertainty_score":0.999867,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0889213252948868,"score_gpt":0.2574665341658127,"score_spread":0.1685452088709259,"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."}}