| 基于改进谱减方法的语音增强研究 |
| |
| 引用本文： | 丁伟,;吴小培.基于改进谱减方法的语音增强研究[J].微机发展,2008(9):98-100. |
| |
| 作者姓名： | 丁伟 ;吴小培 |
| |
| 作者单位： | 安徽大学计算智能与信号处理教育部重点实验室 |
| |
| 基金项目： | 安徽省自然科学基金资助项目（2002kj003）;安徽大学“模式识别”创新基金资助项目 |
| |
| 摘 要： | 由于噪声的影响导致语音信号的质量降低，因此需要对语音信号进行语音增强。语音增强是语音信号处理的前沿领域，其主要目标足从带噪语音中提取纯净的原始语音信号。介绍了实现语音增强方法的原理，利用实验仿真了传统谱减法和改进谱减方法，改进法通过对带噪信号进行参数调整，然后进行频域谱减，实验结果表明改进方法对语音增强效果明显好于传统方法。此外，对传统谱减法和改进谱减法的信噪比分别进行了计算，结果表明改进谱减方法的信噪比相对传统谱减方法有很大提高。
|
| 关 键 词： | 语音增强 改进谱相减法 信噪比 |
Implementation of Speech Enhancement Based on Improved Spectral Subtraction |
| |
| DING Wei,WU Xiao-pei.Implementation of Speech Enhancement Based on Improved Spectral Subtraction[J].Microcomputer Development,2008(9):98-100. |
| |
| Authors: | DING Wei WU Xiao-pei |
| |
| Affiliation: | DING Wei, WU Xiao-pei (Ministry of Education Key Lab. of Intelligent Computing and Signal Processing, Anhui University, Hefei 230039,China) |
| |
| Abstract: | Because of noise,quality of speech signal is degraded,speech enhancement makes very important.Speech enhancement is a long-standing problem in speech processing.Its primary aim is to extract pure speech signal under noise background.In this paper,the fundamental theory of spectral subtraction for noise removal is introduced,then enframe-based spectral subtraction method is proposed.The improved method revises parameters through noise signal.The experiment results show that the improved spectral subtraction method can achieve higher SNR than the traditional method does. |
| |
| Keywords: | speech enhancement spectral subtraction signal to noise ratio |
| 本文献已被 CNKI 维普 等数据库收录！ |
