| 有混响条件下的室内语音定向技术的研究 |
| |
| 引用本文： | 胡跃辉,周康源,陈昕,施俊,周平,黄以华.有混响条件下的室内语音定向技术的研究[J].声学技术,2004,23(1):43-48. |
| |
| 作者姓名： | 胡跃辉 周康源 陈昕 施俊 周平 黄以华 |
| |
| 作者单位： | 中国科学技术大学电子工程与信息科学系,合肥,230027 |
| |
| 摘 要： | 文章讨论了一种新的基于回波分离算法的改进算法，并将其应用于室内有较强混响干扰下的语音定向。算法在有效语音片段检出时，不仅采用了回波估计，还同时考虑了语音特征，最后利用数字波束形成法来计算检出的有效语音片段的方位角，并且，回波估计是在语音短时能量分析的基础上进行的。试验表明，所获得的精度和人耳定向精度相接近。
|
| 关 键 词： | 语音定向 室内混响 回波分离 波束合成 语音处理 声场 语音识别 |
| 文章编号： | 1000-3630(2004)01-0043-06 |
| 收稿时间： | 2002/7/15 0:00:00 |
| 修稿时间： | 2002/9/8 0:00:00 |
Study on speech localization technique in a reverberation room |
| |
| HU Yao-hui,ZHOU Kang-yuan,CHEN Xin,SHI Jun,ZHOU Ping and HUANG Yi-hua.Study on speech localization technique in a reverberation room[J].Technical Acoustics,2004,23(1):43-48. |
| |
| Authors: | HU Yao-hui ZHOU Kang-yuan CHEN Xin SHI Jun ZHOU Ping and HUANG Yi-hua |
| |
| Affiliation: | Department of Electronic Engineering and Information Science, University of Science and Techunology of China, Hefei 230027, China;Department of Electronic Engineering and Information Science, University of Science and Techunology of China, Hefei 230027, China;Department of Electronic Engineering and Information Science, University of Science and Techunology of China, Hefei 230027, China;Department of Electronic Engineering and Information Science, University of Science and Techunology of China, Hefei 230027, China;Department of Electronic Engineering and Information Science, University of Science and Techunology of China, Hefei 230027, China;Department of Electronic Engineering and Information Science, University of Science and Techunology of China, Hefei 230027, China |
| |
| Abstract: | In this paper an improved algorithm based on echo separation is discussed. This algorithm not only uses echo estimation method but also takes the speech characteristics into account when the effective speech segment is separated. A beam-forming method is used to calculate the azimuth of the effective speech segment, and echo estimation is based on a short time energy analysis. The results show that the localization accuracy obtained is very close to that of the human ear. |
| |
| Keywords: | speech localization reverberation echo free beam-forming speech processing |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《声学技术》浏览原始摘要信息 |
|
点击此处可从《声学技术》下载全文 |
|
