首页 | 本学科首页   官方微博 | 高级检索  
     

可区分回波通道变化和双端话音的检测算法研究
引用本文:唐邦友,夏秀渝,何培宇,徐自励. 可区分回波通道变化和双端话音的检测算法研究[J]. 四川大学学报(自然科学版), 2007, 44(1): 73-76
作者姓名:唐邦友  夏秀渝  何培宇  徐自励
作者单位:四川大学电子信息学院,成都,610065;四川大学电子信息学院,成都,610065;四川大学电子信息学院,成都,610065;四川大学电子信息学院,成都,610065
摘    要:自适应声回波对消器中,可区分回波通道变化EPC(Echo Pass Change)和双端话音DT(DoubleTalk)的检测算法非常重要,以确保自适应滤波器在EPC时更新权系数,在DT期间冻结权系数. 提出了两种检测算法:算法1利用自适应滤波器误差信号的平均幅度,并结合误差信号和自适应滤波器输出信号的互相关系数来区分DT和EPC;算法2借助一个辅助滤波器来完成EPC和DT的检测,同时能加快主滤波器的收敛. 仿真结果表明,所提出的检测算法能有效区分EPC和DT,提高了回波对消器性能.

关 键 词:自适应声回波对消器  回波通道变化  双端话音  互相关系数  辅助滤波器
文章编号:0490-6756(2007)01-0073-04
修稿时间:2006-01-23

Study on detection algorithms able to distinguish echo pass change from double-talk
TANG Bang-you,XIA Xiu-yu,HE Pei-yu and XU Zi-li. Study on detection algorithms able to distinguish echo pass change from double-talk[J]. Journal of Sichuan University (Natural Science Edition), 2007, 44(1): 73-76
Authors:TANG Bang-you  XIA Xiu-yu  HE Pei-yu  XU Zi-li
Affiliation:College of Electronics Information,Sichuan University,College of Electronics Information,Sichuan University,College of Electronics Information,Sichuan University,College of Electronics Information,Sichuan University
Abstract:In an adaptive echo canceller,the detection algorithm able to distinguish echo pass change(EPC) from double-talk(DT) is vital to ensure that adaptive filter tap coefficients are updated in case of EPC and frozen during the DT period.Two detection algorithms are proposed.In algorithm 1,adaptive filter error signal's average amplitude is combined with cross-correlation coefficient between error signal and output signal of adaptive filter to distinguish EPC from DT.In algorithm 2,an auxiliary filter is used to detect EPC and DT and able to quicken the convergence of main filter at the same time.The computer simulation results verify that above mentioned algorithms are very useful and efficient in distinguishing EPC from DT and improve the performance of echo canceller.
Keywords:adaptive echo canceller   EPC  DT   cross-correlation coefficient   auxiliary filter
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《四川大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《四川大学学报(自然科学版)》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号