期刊文献+

基于后验信噪比频域迭代算法的语音增强方法

下载PDF
导出
摘要 为了抑制谱减法语音增强时引入的"音乐噪声",采用基于后验信噪比频域迭代算法的语音增强方法。首先,当后验信噪比大于20dB时,对含噪语音采用谱过减法;当后验信噪比小于20dB时,对含噪语音谱线进行衰减处理。为了进一步抑制音乐噪声,对增强语音信号进行多次频域迭代降噪处理。对实验室环境录制的不同输入信噪比条件下的含噪语音信号进行处理,与传统谱减法相比,增强语音信号的信噪比有较大的提高,并且音乐噪声得到很大程度的抑制。
作者 陈紫强
出处 《声学与电子工程》 2007年第3期41-44,共4页 Acoustics and Electronics Engineering
  • 相关文献

参考文献8

  • 1S F Boll.Suppression of acoustic noise in speech using spectral subtraction[J].IEEE Transaction Acoustic,Speech,Signal Processing,1979,27:113-120.
  • 2Y Ephraim,D Malah.Speech enhancement using a minimum mean square error short-time spectral amplitude estimator[J].IEEE Transaction Acoustic Speech signal processing,1984,ASSP-32(12):1109-1121.
  • 3P Lochwood,J Boudy.Experiments with a nonlinear spectral subtractor (NSS),hidden Markov models and projection,for robust recognition in cars[J].Speech Communication,1992,11(5):215-228.
  • 4P Scalart,J Vieira-Filho.Speech enhancement based on a priori signal to noise estimation[C].ICASSP,1996:629-632.
  • 5Udrea R M,Ciochina S.Speech enhancement using Spectral Over-subtraction and residual Noise Reduction[J].IEEE International Symposium on Signals,Circuits and Systems.Proceedings (Cat.No.03EX720),2003,1(7):165-168.
  • 6S Ogata,T Shimamura.Reinforced spectral subtraction method to enhance speech signal[J].Electrical and Electronic Technology,2001,1:242 -245.
  • 7M Marzinzik,B Kollmeier.Speech Pause Detection for Noise Spectrum Estimation by Tracking Power Envelope Dynamics[J].IEEE Transaction Speech Audio Processing,2002,10(2):109-117.
  • 8N Virag.Single Channel Speech enhancement based on masking properties of the human auditory system[J].IEEE Transaction Speech Audio Processing,1999,7(3):126-137.

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部