一种基于自适应谱熵的端点检测改进方法

被引：25

作者：

王琳 ^{[1
,2
]}

李成荣 ^{[1
]}

机构：

[1] 中国科学院自动化研究所

[2] 中国科学院研究生院

来源：

计算机仿真 | 2010年 / 27卷 / 12期

关键词：

语音端点检测; 自适应子带谱熵; 语音识别; 鲁棒性;

D O I：

暂无

中图分类号：

TN912.3 [语音信号处理];

学科分类号：

0711 ;

摘要：

在低信噪比的环境下,为增强与噪声的区分度,提出了一种适应于低信噪比环境的语音端点检测方法。通过改进语音端点检测的特征参数,更好地区分语音信号与噪声信号,提高在低信噪比环境下的端点检测正确率。基于子带谱熵,引入正值常量对基本谱熵参数进行算法改进,得到改良的负谱熵特征,并结合自适应子带选择方法,得到一种新颖的特征参数——自适应子带常量负谱熵。特征在低信噪比的情况下有较强的抗噪能力,并能够准确地检测出语音端点。实验结果表明,不仅快速有效,具有较强的鲁棒性,而且适合低信噪比的语音端点检测。

引用

页码：373 / 375+395 +395

页数：4

共 8 条

[1]

Multi-resolution spectral entropy feature for robust ASR. Hemant Misra,Shajith Ikbal,Sunil Sivadas,Herve Bourlard. IC-ASSP’05 . 2005

[2]

Robust Endpoint DetectionAlgorithm Based on the Adaptive Band-Partitioning Spectral En-tropy in Adverse Environments. Wu Bing-Fei,Wang Kun-Ching. IEEE Transactions on Speechand Audio Processing . 2005

[3]

An Improved Entropy-Based Endpoint Detection algorithm. Chuan Jia,and Bo Xu. ISCSLP‘02 . Aug.22-242002

[4]

Robust entropy-based endpoint detection for speech recognition in noisy environments＂. Jia-lin Shen,Jeih-weih Hung,Lin-shan Lee. International Conference on Spoken Language Processing . 1998

[5]

A Robust Algorithm for Accurate Endpointing of Speech. M H Savoji. Speech Communication . 1989

[6]

Algorithms and Systems for Segmentation and Structure Analysis in Soccer Video. Lie Lu,Hong You,Hong-Jiang Zhang. IEEE International Conference on Multimedia and Expo (ICME 2001) . 2001

[7]

Word Boundary Detection with Mel -Scale Frequency Bank in Noisy Environment. Wu GinDer,Lin ChinTeng. I EEE Transactions on Speech and Audio Processing . 2000

[8]

An Improved Endpoint Detector for Isolated Word Recognition. Lamel L F,Rabiner L R,Rosenberg A E,et al. IEEE Transactions on Acoustics Speech and Signal Processing . 1981

← 1 →