一种基于分带谱熵和谱能量的语音端点检测算法

被引:12
作者
李如玮
鲍长春
机构
[1] 北京工业大学电子信息与控制工程学院
关键词
语音处理; 语音识别; 谱分析; 端点检测; 分带谱熵;
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
语音端点检测的精确度直接影响语音识别的准确度.在噪声环境下,语音端点检测很困难.信噪比下降,语音端点检测的正确率也随之下降,同时,噪声类型的变化影响端点检测的正确率.为此,提出了一种改进的、适合在电话语音城市名识别系统中应用的端点检测算法,并结合分带谱熵和谱能量形成了一个新的特征参数集,利用该参数集进行端点检测,弥补了分别采用分带谱熵和谱能量进行端点检测的缺陷,提高了检测性能.
引用
收藏
页码:920 / 924
页数:5
相关论文
共 8 条
[1]  
Robust endpoint detection algorithm based on the adaptive band-partitioning spectral en- tropy in adverse environments. WU Bing-fei,WANG Kun-ching. IEEE Transactions on Speech and Audio Processing . 2005
[2]  
A novel approach to robust speech endpoint detection in car environments[el. SHENG L S,YANG C H. ICASSP . 2000
[3]  
Robust entropy-based endpoint detection for speech recognition in noisy environments. SHEN J L,HUNG J W,LEE L S. ICSLP . 1998
[4]  
Endpoint detection of speech signal using neural network. AINI H,SALINA A S,LIEW B F. TENCON 2000 Proceed- ings . 2000
[5]  
Endpoint detection of isolated utterances based on a modified teager energy measurement. YING G S,MITCELL C D,JAMIESON L H. IEEE International Conference on Acoustics Speech and Signal Processing Proceedings . 1993
[6]  
A novel chinese continuous speech endpoint detection method based on time domain features of the word structure. HE Su-ning,YU Jue-bang. ICCCSWP . 2002
[7]  
An improved endpoint detect for isolated word recognition. LAMEL L,LAWRENCE R R,ROSENBERG A E,et al. I- TASSP . 1981
[8]  
Robustness and co-operative multi-model man-machine communication applications. HE Su-ning,YU Jue-bang. ICCC- SWP . 2002