Zobrazeno 1 - 6
of 6
pro vyhledávání: '"Huang, Jonathan J"'
We propose an efficient end-to-end convolutional neural network architecture, AclNet, for audio classification. When trained with our data augmentation and regularization, we achieved state-of-the-art performance on the ESC-50 corpus with 85:65% accu
Externí odkaz:
http://arxiv.org/abs/1811.06669
Autor:
Huaxiong Hung1,2 hhuang@yorku.ca, Wylie, Jonathan J.1,2, Mura, Robert M.1,2, Howell, Peter D.2,3 howell@maths.ox.ac.uk
Publikováno v:
SIAM Journal on Applied Mathematics. 2007, Vol. 67 Issue 3, p630-666. 37p. 1 Diagram, 2 Charts, 16 Graphs.
Publikováno v:
Proceedings of SPIE; Nov2011, Issue 1, p795506-795506-5, 5p
Publikováno v:
Bulletin of Mathematical Biology; Jul2011, Vol. 73 Issue 7, p1682-1694, 13p
Publikováno v:
Proceeding of the 17th International Conference: World Wide Web; 4/21/2008, p1155-1156, 2p
Publikováno v:
Proceedings of the 8th ACM/IEEE-CS Joint Conference: Digital Libraries; 6/16/2008, p455-455, 1p