| 基于FT和HHT的语音能量轮郭特征提取 |
| |
| 引用本文： | 刘翠,张歆奕. 基于FT和HHT的语音能量轮郭特征提取[J]. 五邑大学学报(自然科学版), 2014, 0(1): 45-49 |
| |
| 作者姓名： | 刘翠 张歆奕 |
| |
| 作者单位： | 五邑大学信息工程学院,广东江门529020 |
| |
| 摘 要： | 使用FT和HHT分别对男女语音的能量轮廓特征进行提取,并通过聚类性能分析判别两种方法提取的语音能量轮郭特征的有效性.Matlab仿真结果表明,不管是男生分类、女生分类还是男女分类,用HHT提取的语音特征的聚类效果都比FT的效果好,能较好体现不同语音信号的个性信息,有助于提高识别率.
|
| 关 键 词： | 语言识别 傅里叶变换 希尔伯特-黄变换 能量轮廓 |
Energy Contour Features Extraction Based on Fourier Transform and Hilbert- Huang Transform |
| |
| LIU Cui,ZHANG Xin-yi. Energy Contour Features Extraction Based on Fourier Transform and Hilbert- Huang Transform[J]. Journal of Wuyi University(Natural Science Edition), 2014, 0(1): 45-49 |
| |
| Authors: | LIU Cui ZHANG Xin-yi |
| |
| Affiliation: | (School of Information Engineering, Wuyi University, Jiangmen 529020, China) |
| |
| Abstract: | The energy contour features of men＇s and women＇s voice are extracted using the Hilbert-Huang Transform and the Fourier Transform respectively, the effectiveness of the characteristics of the two methods of extraction is determined through the clustering performance analysis. Matlab simulation results show that the clustering effect of the voice features extracted by HHT is better than that by FT. Voice features extracted by HHT can better reflect the personality information of different speakers and can improve recognition rates. |
| |
| Keywords: | vice recognition Fourier Transform Hilbert-Huang Transform energy contour |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
