EXPLOITING SPECTRO-TEMPORAL STRUCTURES USING NMF FOR DNN-BASED  SUPERVISED SPEECH SEPARATION

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 机器人视觉团队

	EXPLOITING SPECTRO-TEMPORAL STRUCTURES USING NMF FOR DNN-BASED SUPERVISED SPEECH SEPARATION
	Shuai, Nie 1; Shan, Liang1 ; Hao, Li 2; XueLiang, Zhang 2; ZhanLei, Yang1 ; WenJu, Liu1 ; LiKe, Dong 3
	2016
会议名称	IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)
会议日期	2016
会议地点	Shang Hai, China
关键词	Speech Separation Deep Neural Network Nonnegative Matrix Factorization Spectro-Temporal Structures
页码	469-473
英文摘要	The targets of speech separation, whether ideal masks or magnitude spectrograms of interest, have prominent spectro-temporal structures. These characteristics are very worthy to be exploited for speech separation, however, they are usually ignored in previous works. In this paper, we use nonnegative matrix factorization (NMF) to exploit the spectro-temporal structures of magnitude spectrograms. With nonnegative constrains, NMF can capture the basis spectra patterns of speech and noise. Then the learned basis spectra are integrated into a deep neural network (DNN) to reconstruct the magnitude spectrograms of speech and noise with their nonnegative linear combination. Using the reconstructed spectrograms, we further explore a discriminative training objective and a joint optimization framework for the proposed model. Systematic experiments show that the proposed model is competitive with the previous methods in monaural speech separation tasks.
收录类别	EI
会议录	IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
会议录出版者	IEEE
语种	英语
内容类型	会议论文
源URL	[http://ir.ia.ac.cn/handle/173211/11023]
专题	自动化研究所_模式识别国家重点实验室_机器人视觉团队
作者单位	1.National Laboratory of Patten Recognition, Institute of Automation, Chinese Academy of Sciences 2.College of Computer Science, Inner Mongolia University 3.Electric Power Research Institute of ShanXi Electric Power Company, China State Grid Corp
推荐引用方式 GB/T 7714	Shuai, Nie,Shan, Liang,Hao, Li,et al. EXPLOITING SPECTRO-TEMPORAL STRUCTURES USING NMF FOR DNN-BASED SUPERVISED SPEECH SEPARATION[C]. 见:IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP). Shang Hai, China. 2016.