| 基于小波包分解和噪声分析的抗噪说话人识别特征参数 |
| |
| 引用本文： | 吴峰燕,李志华.基于小波包分解和噪声分析的抗噪说话人识别特征参数[J].计算机与现代化,2009(1). |
| |
| 作者姓名： | 吴峰燕 李志华 |
| |
| 作者单位： | 河海大学电气工程学院,江苏,南京,210098 |
| |
| 摘 要： | 为提高说话人识别中语音特征参数对噪声的鲁棒性,本文提出在对语音进行小波包分解基础上,分析噪声的特性,在不同子带内进行谱减并设立权重,提出了一种新的语音特征参数多层美尔倒谱系数.仿真实验表明,与MFCC特征参数相比,ML-MFCC在噪声环境下具有更好的抗噪性能和说话人识别率.
|
| 关 键 词： | 多层美尔倒谱系数 小波包分解 说话人识别 |
Robust Feature Based on Noise Analysis and Wavelet Packet for Speaker Recognition |
| |
| WU Feng-yan,LI Zhi-hua.Robust Feature Based on Noise Analysis and Wavelet Packet for Speaker Recognition[J].Computer and Modernization,2009(1). |
| |
| Authors: | WU Feng-yan LI Zhi-hua |
| |
| Affiliation: | College of Electric Engineering;Hohai University;Nanjing 210098;China |
| |
| Abstract: | To improve the performance of speaker recognition in noise environment,a robust feature Multilayer Mel cepstrum coefficient(ML-MFCC) based on noise analysis and wavelet packet for speaker recognition is proposed.Experiments show that ML-MFCC performs better than MFCC in noise environment. |
| |
| Keywords: | MFCC wavelet packet speaker recognition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
