BROWSE
Detail
Conference Paper
Deep Neural Network based Acoustic Model Parameter Reduction Using Manifold Regularized Low Rank Matrix Factorization
Cited 3 time in 
Share 
Share
- Authors
- Hoon Chung, Jeom Ja Kang, Ki Young Park, Sung Joo Lee, Jeon Gue Park
- Issue Date
- 2016-12
- Citation
- Workshop on Spoken Language Technology (SLT) 2016, pp.1-6
- Language
- English
- Type
- Conference Paper
- Abstract
- In this paper, we propose a deep neural network (DNN) model parameter reduction based on manifold regularized low rank matrix factorization to reduce the computational complexity of acoustic model for low resource embedded devices. One of the most common DNN model parameter reduction techniques is truncated singular value decomposition (TSVD). TSVD reduces the number of parameters by approximating a target matrix with a low rank one in terms of minimizing the Euclidean norm. In this work, we questioned whether the Euclidean norm is appropriate as objective function to factorize DNN matrices because DNN is known to learn nonlinear manifold of acoustic features. Therefore, in order to exploit the manifold structure for robust parameter reduction, we propose manifold regularized matrix factorization approach. The proposed method was evaluated on TIMIT phone recognition domain.
- KSP Keywords
- Computational complexity, Deep neural network(DNN), Euclidean norm, Low-rank matrix, Manifold regularized, Manifold structure, Matrix Factorization, Model parameter, Nonlinear manifold, Reduction technique, Target matrix