{"id":"W2005708641","doi":"10.1109/asru.2013.6707742","title":"Hybrid speech recognition with Deep Bidirectional LSTM","year":2013,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":1794,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"TIMIT; Computer science; Word error rate; Speech recognition; Leverage (statistics); Hidden Markov model; Artificial neural network; Recurrent neural network; Artificial intelligence; Deep neural networks; Word (group theory); Vocabulary; Acoustic model; Deep learning; Speech 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007087198,0.0007244415,0.0006105091,0.0005002004,0.0002382685,0.0008719816,0.0009976859,0.0007765848,0.004740981],"category_scores_gemma":[0.001032607,0.00040562,0.000513311,0.0005287207,0.0002291737,0.001396799,0.0008795346,0.0008987238,0.003171199],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005023192,"about_ca_system_score_gemma":0.0004606413,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005112895,"about_ca_topic_score_gemma":0.01006185,"domain_scores_codex":[0.9995766,0.0000819097,0.00002613578,0.0001398884,0.0001209282,0.00005454088],"domain_scores_gemma":[0.9996156,0.0001381731,0.0000241463,0.00007070456,0.0001303614,0.00002097067],"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":[0.0005492383,0.0001879615,0.0008769353,0.0002066932,0.0002151964,0.0002084404,0.0001464562,0.1156329,0.1064973,0.004152012,0.008426363,0.7629004],"study_design_scores_gemma":[0.00001737023,0.00007119899,0.000511224,0.00001341876,0.00003129254,0.00008184098,0.00002960247,0.9671316,0.02629052,0.002837402,0.002958818,0.0000258189],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05921132,0.001228378,0.9147481,0.0003145846,0.0002954403,0.00008470022,0.0009966131,0.01626609,0.006854757],"genre_scores_gemma":[0.6456982,0.0004218286,0.3378663,0.0003298515,0.000102265,0.0001657065,0.001925594,0.0004516861,0.0130386],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005112895,"threshold_uncertainty_score":0.01586014,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01839249538427433,"score_gpt":0.206168135037918,"score_spread":0.1877756396536437,"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."}}