{"id":"W2172097686","doi":"10.1109/icassp.2012.6288863","title":"Understanding how Deep Belief Networks perform acoustic modelling","year":2012,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":270,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Deep belief network; TIMIT; Computer science; Hidden Markov model; Artificial intelligence; Artificial neural network; Similarity (geometry); Feature (linguistics); Pattern recognition (psychology); Speech recognition; Feature vector; Visualization; Representation (politics)","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":[0.0003107303,0.0001076889,0.0001047003,0.00008127094,0.0001972029,0.0002087204,0.000292923,0.00006777603,0.0002264961],"category_scores_gemma":[0.00001537388,0.00009066551,0.00005928898,0.0001767556,0.00001919808,0.0008190137,0.00008380842,0.0001099929,0.00007968699],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001000269,"about_ca_system_score_gemma":0.00001132194,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003550675,"about_ca_topic_score_gemma":0.000005019744,"domain_scores_codex":[0.9990863,0.00002497252,0.0001004734,0.0001727339,0.0001928522,0.0004227101],"domain_scores_gemma":[0.9993851,0.0001457994,0.00003588126,0.0002382273,0.00002791718,0.0001671158],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001430087,0.0003652952,0.002894316,0.00004695772,0.0001393073,0.00001855082,0.002742902,0.1666138,0.0002594939,0.5051228,0.005332418,0.3164498],"study_design_scores_gemma":[0.00008202136,0.00001058798,0.00003501869,0.00000856034,0.00000728945,0.00001630082,0.0002233806,0.9972193,0.0002237922,0.001781028,0.0002466897,0.000146009],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001327514,0.0001043522,0.9750746,0.0004295871,0.0004280631,0.00006323471,1.449858e-7,0.0002116352,0.02236088],"genre_scores_gemma":[0.8830202,0.00004289324,0.115991,0.0003501867,0.0001801361,0.000004635551,7.383667e-7,0.000007900603,0.0004022517],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8816927,"threshold_uncertainty_score":0.3697232,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1542638219167346,"score_gpt":0.2391786981223847,"score_spread":0.08491487620565008,"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."}}