{"id":"W2901145677","doi":"10.48550/arxiv.1811.09678","title":"Speech recognition with quaternion neural networks","year":2018,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Music and Audio Processing","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Quaternion; Context (archaeology); Artificial neural network; Convolutional neural network; Speech recognition; Focus (optics); ENCODE; Representation (politics); Filter (signal processing); Filter bank; Artificial intelligence; Process (computing); TIMIT; Pattern recognition (psychology); Hidden Markov model; Mathematics; Computer vision","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001816594,0.0002691491,0.0002282245,0.0001431689,0.0001848479,0.0002380786,0.001056749,0.0002274258,0.00004851813],"category_scores_gemma":[0.000007317966,0.0002630711,0.0000884353,0.0004213207,0.0001200688,0.0005703134,0.0009719825,0.0004733483,0.00007748179],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008941758,"about_ca_system_score_gemma":0.00008835207,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007704128,"about_ca_topic_score_gemma":0.00003111036,"domain_scores_codex":[0.998378,0.00009394472,0.0001368303,0.0009738904,0.00008535699,0.0003319448],"domain_scores_gemma":[0.998673,0.0000331805,0.0002678194,0.0007067349,0.0001973784,0.000121921],"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":[0.0004690863,0.000456644,0.0164537,0.0006191396,0.0004260618,0.003658303,0.001221447,0.6497012,0.00005993608,0.008915866,0.007404956,0.3106137],"study_design_scores_gemma":[0.0003229307,0.00008966731,0.0005202339,0.0002058008,0.00003709353,0.0000227359,0.00001524657,0.9865701,0.0001727647,0.01141526,0.0001912591,0.0004369194],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3134243,0.00002020436,0.6836153,0.00009355936,0.0004881725,0.0001177125,0.000001910077,0.0002077433,0.002031039],"genre_scores_gemma":[0.9935763,0.00003748925,0.005140145,0.0003232445,0.0003171099,5.40603e-7,0.00002211289,0.00001688553,0.0005661385],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.680152,"threshold_uncertainty_score":0.9999822,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08138138747732042,"score_gpt":0.1814880062847343,"score_spread":0.1001066188074139,"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."}}