{"id":"W2964250984","doi":"10.1109/icassp.2018.8461624","title":"An Experimental Analysis of the Power Consumption of Convolutional Neural Networks for Keyword Spotting","year":2018,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":48,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Keyword spotting; Computer science; Proxy (statistics); Convolutional neural network; Inference; Predictive power; Artificial intelligence; Footprint; Memory footprint; Spotting; Energy consumption; Predictive modelling; Artificial neural network; Machine learning; Data mining","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":[0.0004839558,0.0007497965,0.0003446483,0.0003966083,0.0002744008,0.0004082591,0.0009878207,0.0003766329,0.004005234],"category_scores_gemma":[0.003810687,0.0002010406,0.0002119565,0.0006084872,0.0003422737,0.001163138,0.0002794109,0.0005676795,0.0005765927],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007910116,"about_ca_system_score_gemma":0.0004070821,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004573845,"about_ca_topic_score_gemma":0.0063469,"domain_scores_codex":[0.999589,0.00006981876,0.00003958421,0.00008682541,0.0001236884,0.00009111885],"domain_scores_gemma":[0.9979495,0.001228085,0.0001282332,0.0002496764,0.0003803914,0.00006407405],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.005722887,0.00118862,0.0138539,0.001138431,0.0003196879,0.0009630176,0.0003271722,0.4499207,0.1863814,0.00388344,0.009903224,0.3263976],"study_design_scores_gemma":[0.00009017708,0.001138979,0.007800106,0.00003205856,0.00008656117,0.0002318125,0.0001336325,0.8593622,0.1271059,0.00138811,0.002601097,0.00002932092],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9756964,0.0003968735,0.0174447,0.0002449809,0.0000837397,0.00005146847,0.0006574754,0.001325751,0.004098461],"genre_scores_gemma":[0.9936935,0.000120656,0.004203897,0.00003452231,0.000008279668,0.00003338953,0.0003814003,0.00008599846,0.001438426],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.004573845,"threshold_uncertainty_score":0.01339883,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03467712234277139,"score_gpt":0.2989291442651436,"score_spread":0.2642520219223722,"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."}}