| 噪声环境下说话人识别性能的研究 |
| |
| 引用本文： | 张飞云,蔡子亮,盛胜我.噪声环境下说话人识别性能的研究[J].电声技术,2007,31(6):41-43. |
| |
| 作者姓名： | 张飞云 蔡子亮 盛胜我 |
| |
| 作者单位： | 1. 许昌学院,电气信息工程学院,河南,许昌,461000
2. 同济大学,声学研究所,上海,200092 |
| |
| 摘 要： | 为了提高噪声环境下说话人识别系统的识别性能,将基于听觉掩蔽效应的语音增强技术作为预处理器,对语音信号首先进行降噪处理,提高输入信号的信噪比。实验证明,经过降噪处理的语音信号送入说话人识别系统,提高了系统的识别性能。
|
| 关 键 词： | 说话人识别 高斯混合模型 特征提取 掩蔽阈值 语音增强 |
| 文章编号： | 1002-8684（2007）06-0041-03 |
| 修稿时间： | 2007-02-12 |
Research on Speaker Recognition Under Noisy Environment |
| |
| ZHANG Fei-yun,CAI Zi-liang,SHENG Sheng-wo.Research on Speaker Recognition Under Noisy Environment[J].Audio Engineering,2007,31(6):41-43. |
| |
| Authors: | ZHANG Fei-yun CAI Zi-liang SHENG Sheng-wo |
| |
| Institution: | 1. Xuchang University, Xuchang Henan 461000, China; 2. Tongji University, Shanghai 200092, China |
| |
| Abstract: | Speech enhancement method based on masking properties of the human auditory system is used to reduce the white noise in the front-end. Experimental results show that the performance of speaker recognition system is improved with the processed speech signal in the noise environments. |
| |
| Keywords: | speaker recognition gaussian mixture model feature extract masking properties speech enhancement |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
