{"id":"W2964187693","doi":"10.1109/icassp.2018.8462688","title":"Deep Residual Learning for Small-Footprint Keyword Spotting","year":2018,"lang":"en","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":227,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Keyword spotting; Computer science; Residual; Convolutional neural network; Deep learning; Footprint; Memory footprint; Benchmark (surveying); Artificial intelligence; Spotting; Machine learning; Speech recognition; Algorithm; Cartography","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":[0.0004389965,0.00008724374,0.00009872248,0.00008103303,0.0002403683,0.0001639684,0.0003982715,0.0000459548,0.0001751225],"category_scores_gemma":[0.0003452814,0.00007699701,0.00005884938,0.0001532757,0.00003492016,0.00008730525,0.0001409629,0.00007543112,0.0002731074],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001766652,"about_ca_system_score_gemma":0.00002546378,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001528034,"about_ca_topic_score_gemma":0.00004936984,"domain_scores_codex":[0.9991183,0.00004479012,0.0001621718,0.0002944797,0.000117772,0.0002625301],"domain_scores_gemma":[0.9992498,0.0002599815,0.00005304314,0.0002178709,0.0001445885,0.00007471046],"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":[0.000007756544,0.00001937345,0.0004424587,0.000005756944,0.00001124766,0.000002633699,0.0003355568,0.000006301837,0.0009498247,0.02059416,0.000223745,0.9774012],"study_design_scores_gemma":[0.001205193,0.0006628312,0.008610711,0.00007027026,0.00002134848,0.00006798679,0.0008102995,0.514668,0.3284133,0.01670033,0.1278835,0.0008861953],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0326277,0.00000834525,0.9115483,0.0007730432,0.000228317,0.0001174358,8.836763e-8,0.0003276812,0.05436911],"genre_scores_gemma":[0.3253387,0.000002344914,0.6703811,0.0004422787,0.000326429,0.00002004643,7.250401e-7,0.000009640082,0.003478793],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.976515,"threshold_uncertainty_score":0.3510334,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04154259823274774,"score_gpt":0.2640082849263389,"score_spread":0.2224656866935912,"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."}}