| 部分HHT和ICA的混合语音增强算法 |
| |
| 引用本文： | 王晓建,王玲,彭启琮.部分HHT和ICA的混合语音增强算法[J].电子科技大学学报(社会科学版),2008(Z1). |
| |
| 作者姓名： | 王晓建 王玲 彭启琮 |
| |
| 作者单位： | 电子科技大学通信与信息工程学院; |
| |
| 摘 要： | 根据语音信号中噪声特性的不同,该文将其中的白噪声和色噪声分别进行处理以达到较好的语音增强效果。对信号进行不完全希尔伯特-黄变换,从中得到含有白噪声的分量并进行处理,降低白噪声在整个语音信号中的能量;利用独立分量分析的方法将目标语音与色噪声进行分离,得到最终的目标语音信号。实验结果证明,该混合算法有很好的性能和降噪效果,提升后的语音信号几乎完全消除了色噪声的影响,并相对白噪声有很好的抑制,对于实际和理论模型的不完全吻合也有较好的鲁棒性。
|
| 关 键 词： | 高阶矩分析 希尔伯特-黄变换 独立分量分析 语音增强 |
A Hybrid Method for Speech Enhancement Based on HHT and ICA |
| |
| WANG Xiao-jian,WANG Lin, PENG Qi-cong.A Hybrid Method for Speech Enhancement Based on HHT and ICA[J].Journal of University of Electronic Science and Technology of China(Social Sciences Edition),2008(Z1). |
| |
| Authors: | WANG Xiao-jian WANG Lin PENG Qi-cong |
| |
| Institution: | WANG Xiao-jian,WANG Lin,, PENG Qi-cong (School of Communication , Information Engineering,University of Electronic Science , Technology of China Chengdu 610054) |
| |
| Abstract: | According to the distinct properties of different noise components, this paper classifies the noises into two categories and processed separately to get a good speech enhancement performance: the white noise is processed by partial Hilbert-Huang transform (HHT), and the colored noises is separated from onset speech by independent component analysis (ICA). The experimental results show that the method has a good performance: almost no colored noise presents in the speech signal after enhancement and the whit... |
| |
| Keywords: | high-order statistic analysis Hilbert-Huang transform independent component analysis speech enhancement |
| 本文献已被 CNKI 等数据库收录！ |
|
