后置维纳滤波和可调波束成形器的语音信号增强
|
摘要
针对自适应波束形成滤器会带来误差噪声,提出一种两级滤波器结构的语音增强方法。第1级由一个可调滤波器与4个麦克风阵列的求和波束形成器组成。通过一个控制信号控制波束形成滤波器。第2级是一个维纳滤波器,通过两个相邻的主输出定向光束之间的互相关实现信号功率谱估计,该估计是基于两个相邻定向光束的噪声输出是来自两个独立的噪声源,语音输出源来自同一个信号源。仿真结果表明,算法可以提高信噪比约6 db。
Due to the error noise produced by adaptive beam-forming filter,a two-stage filter structure for speech enhancement is introduced. The first stage of the proposed filter is a control signal that is used to control beam-forming filter,and it is composed of an adjustable filter and sum beam-former with four-microphone array. The second stage of the proposed filter is a Wiener filter,in this stage the signal power spectrum estimation is realized by cross-correlation of two adjacent primary output directional beams. This estimation is based on the assumption that the noise output of two adjacent directional beams comes from two independent noise source and the speech output comes from the same signal source. The simulation experiment shows that the proposed algorithm can improve the signal-noise-ratio(SNR)for about 6 db.
Due to the error noise produced by adaptive beam-forming filter,a two-stage filter structure for speech enhancement is introduced. The first stage of the proposed filter is a control signal that is used to control beam-forming filter,and it is composed of an adjustable filter and sum beam-former with four-microphone array. The second stage of the proposed filter is a Wiener filter,in this stage the signal power spectrum estimation is realized by cross-correlation of two adjacent primary output directional beams. This estimation is based on the assumption that the noise output of two adjacent directional beams comes from two independent noise source and the speech output comes from the same signal source. The simulation experiment shows that the proposed algorithm can improve the signal-noise-ratio(SNR)for about 6 db.
引文
[1]陈婵娟.移动通信语音质量评估系统的设计与实现[D].厦门:厦门大学,2011.
[2]李克粉,王直.改进的小波阈值去噪在语音识别中的应用[J].计算机技术与发展,2013,23(5):231-234.
[3]许佳龙,张一闻,刘建平.基于麦克阵列的信号采集处理系统设计[J].电子科技,2011,24(9):124-127.
[4]钟强,黄青华.基于球麦克风阵列的多声源定位方法[J].计算机工程与应用,2012,48(5):149-152.
[5]周述畅,宋亚男,吴光波.基于麦克风阵列的语音增强研究[J].计算机技术与发展,2012,22(7):204-206.
[6]王勇,刘颖,刘建平.一种基于麦克风阵列的声源定位算法研究[J].现代电子技术,2011,34(19):61-64.
[7]FROST O L.An algorithm for linealy constrained adaptive array processing[J].Proceedings of the IEEE,1972,60(8):926-935.
[8]GRENIER Y.A microphone array for car environments[J].Speech Communication,1993,12(1):25-39.
[9]GOULDING M,BIRD J S.Speech enhancement for mobile telephony[J].IEEE Trans.Vehicular Technology,1990,39(4):316-326.
[10]AFFES S,GRENIE Y.The adaptive beam-formers for speech acquisition in cars[J].International Conference on Signal Processing Applications and Technology,1994(10):154-159.
[11]张丽艳.复杂环境下麦克风阵列语音增强方法研究[D].大连:大连理工大学,2009.
[12]BITZER J,SIMMER K U,KAMMEYER K D.Theoretical noise reduction limits of the generalized sidelobe canceller(GSC)for speech enhancement[J].IEEE Int ernational Conference on Acoustics,1999,5(171):2965-2968.
[13]ZELINSKI R.A microphone array with adaptive post-filtering for noise reduction in reverberant rooms[J].IEEE International Conference on Acoustics,Speech and Signal Processing,1988(5):2578-2581.
[14]MEYER J,SIMMER K U.Multichannel speech enhancement in car environment using wiener filtering and spectral subtraction[J].IEEE International Conference on Acoustics,Speech and Signal Processing,1997(2):1167-1170.
[15]MCCOWAN I A,BOURLAND H.Microphone array post-filter for diffuse noise field[J].IEEE International Conference on Acoustics,Speech and Signal Processing,2002(1):905-908.
[16]KAJALA M,HAMALAINEN M.Filter-and-sum beam-former with adjustable filter characteristics[J].IEEE International Conference on Acoustics,Speech and Signal Processing,2001(5):2917-2920.
[2]李克粉,王直.改进的小波阈值去噪在语音识别中的应用[J].计算机技术与发展,2013,23(5):231-234.
[3]许佳龙,张一闻,刘建平.基于麦克阵列的信号采集处理系统设计[J].电子科技,2011,24(9):124-127.
[4]钟强,黄青华.基于球麦克风阵列的多声源定位方法[J].计算机工程与应用,2012,48(5):149-152.
[5]周述畅,宋亚男,吴光波.基于麦克风阵列的语音增强研究[J].计算机技术与发展,2012,22(7):204-206.
[6]王勇,刘颖,刘建平.一种基于麦克风阵列的声源定位算法研究[J].现代电子技术,2011,34(19):61-64.
[7]FROST O L.An algorithm for linealy constrained adaptive array processing[J].Proceedings of the IEEE,1972,60(8):926-935.
[8]GRENIER Y.A microphone array for car environments[J].Speech Communication,1993,12(1):25-39.
[9]GOULDING M,BIRD J S.Speech enhancement for mobile telephony[J].IEEE Trans.Vehicular Technology,1990,39(4):316-326.
[10]AFFES S,GRENIE Y.The adaptive beam-formers for speech acquisition in cars[J].International Conference on Signal Processing Applications and Technology,1994(10):154-159.
[11]张丽艳.复杂环境下麦克风阵列语音增强方法研究[D].大连:大连理工大学,2009.
[12]BITZER J,SIMMER K U,KAMMEYER K D.Theoretical noise reduction limits of the generalized sidelobe canceller(GSC)for speech enhancement[J].IEEE Int ernational Conference on Acoustics,1999,5(171):2965-2968.
[13]ZELINSKI R.A microphone array with adaptive post-filtering for noise reduction in reverberant rooms[J].IEEE International Conference on Acoustics,Speech and Signal Processing,1988(5):2578-2581.
[14]MEYER J,SIMMER K U.Multichannel speech enhancement in car environment using wiener filtering and spectral subtraction[J].IEEE International Conference on Acoustics,Speech and Signal Processing,1997(2):1167-1170.
[15]MCCOWAN I A,BOURLAND H.Microphone array post-filter for diffuse noise field[J].IEEE International Conference on Acoustics,Speech and Signal Processing,2002(1):905-908.
[16]KAJALA M,HAMALAINEN M.Filter-and-sum beam-former with adjustable filter characteristics[J].IEEE International Conference on Acoustics,Speech and Signal Processing,2001(5):2917-2920.