《计算机应用研究》|Application Research of Computers

基于可调Q-因子小波变换的语音增强算法

Speech enhancement algorithm based on tunable Q-factor wavelet transform

免费全文下载 (已被下载 次)  
获取PDF全文
作者 殷明,孔冉冉
机构 合肥工业大学 数学学院,合肥 230009
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2014)11-3316-04
DOI 10.3969/j.issn.1001-3695.2014.11.026
摘要 针对语音增强算法中传统的小波阈值法的局限性,提出一种基于可调Q-因子小波变换和清浊音分离的语音增强算法。首先用过零率和短时能量法判别清音和浊音;然后在可调Q-因子小波变换下,对清、浊音采用不同的阈值处理,在不同尺度上,分别结合系数能量和噪声方差得到的阈值作为清音和浊音的阈值确定准则;再利用改进的阈值函数分别处理清音和浊音的小波系数,估计出不含噪声的系数;最后进行小波逆变换,得到抑制了噪声的语音信号。对含有高斯白噪声和有色噪声的语音进行仿真实验,结果表明:与目前许多经典的去噪方法相比,该方法在去噪效果和提高语音可懂度方面均有一定的改善。
关键词 可调Q-因子小波变换;语音增强;清浊音分离;Donoho阈值;阈值函数
基金项目 合肥工业大学博士专项科研基金资助项目(2012HGBZ0653)
安徽省自然科学基金资助项目(1308085MA09)
安徽省教育厅基金资助项目(2013AJZR0039)
本文URL http://www.arocmag.com/article/01-2014-11-026.html
英文标题 Speech enhancement algorithm based on tunable Q-factor wavelet transform
作者英文名 YIN Ming, KONG Ran-ran
机构英文名 School of Mathematics, Hefei University of Technology, Hefei 230009, China
英文摘要 Aiming at the limitations of methods on speech enhancement by traditional threshold methods in wavelet domain, this paper proposed a new speech enhancement algorithm based on the tunable Q-factor wavelet transform and separation of voiced signal and unvoiced signal. Firstly, it separated voiced signal and unvoiced signal with zero-crossing ratio and short-time energy. The adaptive threshold values combined the energy of coefficients and the variance of noise in different scales, respectively. Then it applied the improved Donoho threshold value and threshold function to process wavelet coefficients of voiced signal and unvoiced signal, and estimated the original coefficients from noisy coefficients. Lastly, it used the inverse transform to obtain the original speech signal which the noise was removed. Comparing with the other current classical algorithms, experimental results show that the modified algorithm improves the effect of de-noising and speech intelligibility in white Gaussian noise and colored noise background.
英文关键词 tunable Q-factor wavelet transform(TQWT); speech enhancement; separation of voiced signal and unvoiced signal; Donoho threshold; threshold function
参考文献 查看稿件参考文献
  [1] 张雪英. 数字语音处理及MATLAB仿真[M] . 北京:电子工业出版社, 2010:207.
[2] SANSAM T F, SHAHNAZ C. Noisy speech enhancement based on an adaptive threshold and a modified hard thresholding function in wavelet packet domain[J] . Digital Signal Processing, 2013, 23(3):941-951.
[3] 王慧琴. 小波分析与应用[M] . 北京:北京邮电大学出版社, 2011:19-20.
[4] SELESNICK I W. Wavelet transform with tunable Q-factor[J] . Trans on Signal Processing, 2011, 59(8):3560-3575.
[5] 郭海燕. 基于小波变换的语音增强算法[D] . 秦皇岛:燕山大学, 2012.
[6] DONOHO D L, JOHNSTONE I M. Ideal spatial adaptation via wavelet shrinkage[J] . Biometrika, 1994, 81(3):425-455.
[7] JOHNSTONE I M, SILVERMAN B W. Wavelet threshold estimators for data with correlated noise[J] . Journal of the Royal Statistical Society Series B, 1997, 59(2):319-351.
[8] 潘泉, 张磊, 孟晋丽, 等. 小波滤波方法及应用[M] . 北京:清华大学出版社, 2005.
[9] AKHAEE M A, AMERI A, MARVASTI F A. Speech enhancement by adaptive noise cancellation in the wavelet domain[C] //Proc of the 5th International Conference on Information, Communications and Signal Processing. 2005:719-723.
[10] BACCHELLI S, PAPI S. Filtered wavelet thresholding methods[C] //Proc of the 10th International Congress on Computational and Applied Mathematics. 2004:39-52.
[11] GHANBARI Y, MOLLAEI M R K. A new approach for speech enhancement based on the adaptive thresholding of the wavelet packets[J] . Speech Communication, 2006, 48(8):927-940.
[12] JOHNSON M T, YUAN Xiao-long, REN Yao. Speech signal enhancement through adaptive wavelet thresholding[J] . Speech Communication, 2007, 49(2):123-133.
[13] 段其昌, 邓玉娟, 应泽贵. 基于改进阈值函数的小波包语音增强算法的研究[J] . 通信技术, 2009, 42(5):86-88.
[14] WANG Chun-li, ZHANG Chen-lei, ZHANG Peng-tu. Denoising algorithm based on wavelet adaptive threshold[J] . Physics Procedia, 2012, 24(A):678-685.
[15] 张君昌, 叶珍, 李艳艳. 一种基于清浊音分离的动态阈值小波去噪方法[J] . 计算机工程与应用, 2011, 47(12):133-136.
[16] 戴维, 于盛林, 孙栓. 基于Contourlet变换自适应阈值的图像去噪算法[J] . 电子学报, 2007, 35(10):1939-1943.
[17] 任永梅, 张雪英, 贾海蓉. 一种新阈值函数的小波包语音增强算法[J] . 计算机应用研究, 2013, 30(1):114- 137.
[18] JIA Hai-rong, ZHANG Xue-ying, BAI Jing. A continuous differentiable wavelet threshold function for speech enhancement[J] . Journal of Central South University, 2013, 20(8):2219-2225.
[19] 徐文博, 武晓春, 邢建平. 一种新的小波阈值去噪算法[J] . 兰州交通大学学报, 2012, 31(3):120-124.
[20] HADHAMI I, BOUZID A. Speech denoising based on empirical mode decomposition and improved thresholding[C] //Proc of NOLISP. 2013:200-207.
[21] BACHU R G, KOPPARTHI S, ADAPA B K, et al. Separation of voiced and unvoiced speech signals using energy and zero crossing rate[C] //Proc of ASEE Regional Conference. 2008.
收稿日期 2013/9/18
修回日期 2013/11/6
页码 3316-3319,3323
中图分类号 TN912.35;TP301.6
文献标志码 A