An Improved Spectral Subtraction Method Based on Modified MCRA Noise Estimate and Magnitude Compensation

被引:0
作者
Wang, Xing [1 ]
Wei, Jie [1 ]
Zhong, Xiaofeng
机构
[1] Beijing Jiaotong Univ, Sch Elect & Informat Engn, Beijing, Peoples R China
来源
2013 IEEE INTERNATIONAL SYMPOSIUM ON INDUSTRIAL ELECTRONICS (ISIE) | 2013年
关键词
spectral subtraction; noise estimate; magnitude compensation; SPEECH ENHANCEMENT;
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Conventional spectral subtraction algorithm is limited by its stationary noise assumption. However, most of the nonstationary noises are encountered in real world. In this paper, a novel refined spectral subtraction algorithm for speech enhancement is proposed. The algorithm based on both MCRA (Minima Controlled Recursive Averaging) noise estimate algorithm and magnitude compensation algorithm, which could keep residual noise and speech distortion under a lower level
引用
收藏
页数:4
相关论文
共 10 条
[1]  
[Anonymous], P IEEE INT C AC SPEE
[2]   SUPPRESSION OF ACOUSTIC NOISE IN SPEECH USING SPECTRAL SUBTRACTION [J].
BOLL, SF .
IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1979, 27 (02) :113-120
[3]  
Bouchikhi El, 2012, IEEE INT S IND EL IS, P1228
[4]   Enhancement of Speech Recognitions for Control Automation Using an Intelligent Particle Swarm Optimization [J].
Chan, Kit Yan ;
Yiu, Cedric K. F. ;
Dillon, Tharam S. ;
Nordholm, Sven ;
Ling, Sai Ho .
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2012, 8 (04) :869-879
[5]   Noise estimation by minima controlled recursive averaging for robust speech enhancement [J].
Cohen, I ;
Berdugo, B .
IEEE SIGNAL PROCESSING LETTERS, 2002, 9 (01) :12-15
[6]  
Hu Y., 2001, P IEEE INT C AC SPEE, P673
[7]   Subjective comparison and evaluation of speech enhancement algorithms [J].
Hu, Yi ;
Loizou, Philipos C. .
SPEECH COMMUNICATION, 2007, 49 (7-8) :588-601
[8]  
Kamath S., 2002, P ICASSP 2002 ORL FL
[9]  
Li Chao, ICASSP2011, P4760
[10]  
Wu Gin-der, 2009, IEEE INT S IND EL, P800