Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (33): 214-216.

• 工程与应用 • Previous Articles     Next Articles

Improved speech separation algorithm based on EASI

MENG Dong-xia,MA Jian-fen,QIAO Yong-feng   

  1. 1.College of Sciences,Taiyuan University of Technology,Taiyuan 030024,China
    2.Institute of Computer and Software,Taiyuan University of Technology,Taiyuan 030024,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-21 Published:2007-11-21
  • Contact: MENG Dong-xia

一种改进的基于EASI的语音分离算法

孟东霞,马建芬,乔永凤   

  1. 1.太原理工大学 理学院,太原 030024
    2.太原理工大学 计算机与软件学院,太原 030024
  • 通讯作者: 孟东霞

Abstract: Independent component analysis is a kind of widely used signal processing method developed in recent years because it does not need to obtain the prior knowledge of signals.In this paper we first have introduced the basic method of the ICA and EASI algorithm.Then we have provided an improved one used in speech separation based on neural network.

Key words: speech separation, independent component analysis, EASI algorithm, neural network

摘要: 独立分量分析是近年来发展起来的一门新的数字信号处理方法,因为不需要知道信号的先验信息而得到广泛应用。论文简单介绍了ICA的原理及EASI算法,并根据神经网络理论提出一种改进的EASI语音分离算法。

关键词: 语音分离, 独立分量分析, EASI算法, 神经网络