摘要
用扩展联合对角化(JADE)算法对混合语音信号进行盲源分离,并对分离前后的语音信号的波形和频谱进行了分析比较.实验表明,JADE算法在混合语音信号的分离中是有效的.
The joint approximative diagonalization of eigenmatrix is introduced in this paper and some experiments on blind source separation of mixed voice signals are made. The waveform and spectrum of the speech signals before and after separation are compared and analyzed. Experiments show that JADE is an effective algorithm for the blind source separation of the mixed speech signals.
出处
《五邑大学学报(自然科学版)》
CAS
2008年第4期10-15,共6页
Journal of Wuyi University(Natural Science Edition)
关键词
盲信号处理
盲源分离
扩展联合对角化算法
语音信号分离
blind signal processing
blind source separation (BSS)
JADE algorithm
speech signal separation