標題: Variational Recurrent Neural Networks for Speech Separation
作者: Chien, Jen-Tzung
Kuo, Kuan-Ting
電機工程學系
Department of Electrical and Computer Engineering
關鍵字: recurrent neural network;variational learning;speech separation
公開日期: 1-Jan-2017
摘要: We present a new stochastic learning machine for speech separation based on the variational recurrent neural network (VRNN). This VRNN is constructed from the perspectives of generative stochastic network and variational auto-encoder. The idea is to faithfully characterize the randomness of hidden state of a recurrent neural network through variational learning. The neural parameters under this latent variable model are estimated by maximizing the variational lower bound of log marginal likelihood. An inference network driven by the variational distribution is trained from a set of mixed signals and the associated source targets. A novel supervised VRNN is developed for speech separation. The proposed VRNN provides a stochastic point of view which accommodates the uncertainty in hidden states and facilitates the analysis of model construction. The masking function is further employed in network outputs for speech separation. The benefit of using VRNN is demonstrated by the experiments on monaural speech separation.
URI: http://dx.doi.org/10.21437/Interspeech.2017-832
http://hdl.handle.net/11536/150995
ISSN: 2308-457X
DOI: 10.21437/Interspeech.2017-832
期刊: 18TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2017), VOLS 1-6: SITUATED INTERACTION
起始頁: 1193
結束頁: 1197
Appears in Collections:Conferences Paper