| 基于交叉熵顺序统计滤波的语音端点检测算法 |
| |
| 引用本文： | 钱彦旻,刘加.基于交叉熵顺序统计滤波的语音端点检测算法[J].清华大学学报(自然科学版),2009(10). |
| |
| 作者姓名： | 钱彦旻 刘加 |
| |
| 作者单位： | 清华大学电子工程系清华信息科学与技术国家实验室; |
| |
| 基金项目： | 国家自然科学基金资助项目(60776800);;国家“八六三”高技术项目(2006AA0101012007AA04Z2232008AA02Z414) |
| |
| 摘 要： | 为提高语音端点检测在强噪声环境下的准确率,提出了一种基于交叉熵顺序统计滤波(OSF)的语音端点检测算法。该算法以子带交叉熵为语音/非语音的区分特征,首先将每帧语音的频谱划分成若干个子带,估计出每个子带能量与背景噪声之间的交叉熵,然后把相继若干帧的子带能量交叉熵经过一组顺序统计滤波器,最后根据各帧交叉熵的值对输入的语音进行分类。实验结果表明:该算法能够有效地区分语音和非语音。特别是在强噪声环境下依然能够保持很高的检测率,具有鲁棒性。通过实验结果比较,该算法在性能上优于最近提出的基于能量顺序统计滤波和单纯交叉熵判别的两种方法。
|
| 关 键 词： | 语音信号处理 端点检测 子带交叉熵 顺序统计滤波（OSF） |
Voice activity detection algorithm based on cross-entropy order statistics filter |
| |
| QIAN Yanmin,LIU Jia.Voice activity detection algorithm based on cross-entropy order statistics filter[J].Journal of Tsinghua University(Science and Technology),2009(10). |
| |
| Authors: | QIAN Yanmin LIU Jia |
| |
| Institution: | Tsinghua National Laboratory for Information Science and Technology;Department of Electronic Engineering;Tsinghua University;Beijing 100084;China |
| |
| Abstract: | Voice activity detection in strong noise environments is improved by an algorithm based on the cross-entropy with an order statistics filter(OSF).The algorithm makes use of the sub-band cross-entropy as the speech/non-speech discrimination feature.The analyses first divides the speech spectrum into several sub-bands and then estimates the cross-entropy between the speech signal and the non-speech signal.An order statistics filter is applied to a sequence of the sub-band cross-entropies to obtain the cross-e... |
| |
| Keywords: | |
| 本文献已被 CNKI 等数据库收录！ |
