| 一种基于信号子空间和听觉掩蔽效应的语音增强方法 |
| |
| 引用本文： | 徐望,丁琦,王炳锡.一种基于信号子空间和听觉掩蔽效应的语音增强方法[J].电声技术,2003(12):41-44. |
| |
| 作者姓名： | 徐望 丁琦 王炳锡 |
| |
| 作者单位： | 中国人民解放军信息工程大学,信息工程学院,河南,郑州,450002 |
| |
| 摘 要： | 采用子空间分解法对噪声语音进行信号增强，并用基于Johnston听觉掩蔽模型的感知滤波器对增强后的信号频谱进行平滑以抑制背景音乐噪声。几种噪声背景下对增强语音的客观测试表明，与传统的信号子空间分解法相比，基于信号子空间和听觉掩蔽效应的方法可有效地减少语音信号的失真度。
|
| 关 键 词： | 语音增强 信号子空间 听觉掩蔽效应 感知滤波器 |
| 修稿时间： | 2003年6月7日 |
A method for speech enhancement based on signal subspace and hearing masking effect |
| |
| XIU-Wang,DING-Qi,et.al..A method for speech enhancement based on signal subspace and hearing masking effect[J].Audio Engineering,2003(12):41-44. |
| |
| Authors: | XIU-Wang DING-Qi etal |
| |
| Abstract: | In this paper,a method of speech enhancement is proposed by signal subspace approach.By employing an auditory post -filter based on hearing masking properties to smooth the enhanced speech spectral,the musical noise can be suppressed.The algorithm is tested against different noise background.Compared with the conventional signal subspace based method,objective measurement demonstrates that the method can be used to reduce the distortion of the speech signal. |
| |
| Keywords: | speech enhancement signal subspace hearing masking properties auditory post-filter |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
