{"id":"W4408051557","doi":"10.3390/electronics14050973","title":"Efficient Ensemble of Deep Neural Networks for Multimodal Punctuation Restoration and the Spontaneous Informal Speech Dataset","year":2025,"lang":"en","type":"article","venue":"Electronics","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Punctuation; Computer science; Artificial neural network; Deep neural networks; Speech recognition; Recurrent neural network; Artificial intelligence; Natural language processing","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":[],"consensus_categories":[],"category_scores_codex":[0.0004652913,0.00006968827,0.00009792216,0.00004097561,0.0001077909,0.00005640601,0.0002618011,0.00004134294,4.417773e-7],"category_scores_gemma":[0.00009512735,0.00005155445,0.0000237956,0.0001326723,0.00002780685,0.00008627045,0.00009693844,0.0001050111,2.513056e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003972871,"about_ca_system_score_gemma":0.00005866296,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001300069,"about_ca_topic_score_gemma":0.0001717824,"domain_scores_codex":[0.9993183,0.00003569034,0.0001937419,0.0001470262,0.0001125421,0.0001926775],"domain_scores_gemma":[0.9993235,0.0002265583,0.00007550351,0.0003026569,0.0000565413,0.00001521773],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001329758,0.0000147081,0.00001366565,0.00001358442,0.00001108194,0.000001623854,0.000331137,0.6804839,0.0001236392,0.06858961,0.0001210145,0.2501631],"study_design_scores_gemma":[0.0006625121,0.000052233,0.00003724235,0.00000313386,0.00001000205,0.0000214631,0.000008544725,0.9971806,0.0003483066,0.000859978,0.0007695036,0.00004650725],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08551624,0.0008400633,0.9126612,0.0004233994,0.0001483406,0.0003378809,0.000003889428,0.00001925411,0.00004974869],"genre_scores_gemma":[0.9836969,0.00001565839,0.01594389,0.0002060874,0.00003218072,0.00001637541,0.0000669704,0.000002610485,0.0000192739],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8981807,"threshold_uncertainty_score":0.2102329,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008051341541456172,"score_gpt":0.2394607650231359,"score_spread":0.2314094234816797,"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."}}