Speech Endpoint Detection Based on Improved Cepstral Mean Subtraction

摘要：

This paper presents a novel endpoint detection method based on Cepstral Mean Subtraction (CMS) for robust and accurate speech recognition in noisy environments. The improved method based on CMS applies Hidden Markov Model (HMM) to do two-step classification for better performance, using the optimal spectral feature subset extracted according to the rule of minimum conditional entropy. In addition, to reduce misrecognition due to the similarity between unvoiced sound and white noise in cepstral feature, we apply weighted smoothing algorithm as a solution. Experiment results show that the proposed method outperforms the conventional approaches in both robustness and accuracy relatively.

关键词： Endpoint detection CMS Conditional Information Entropy Weighted smoothing

作者: Du Feifei Huang Qizhi Wei Chengyuan Wang Bo

作者单位: Academy of Military Transportation, Tianjin 300161, China 77th Unit of Force 95971 Command

会议类型: 国际会议

会议名称: 2012 International Conference on Intelligent System Design and Engineering Applications(2012年智能系统设计与工程应用国际会议 ISDEA 2012)

会议地点: 三亚

会议语种:英文

页码: 1121-1124

在线出版日期: 2012-01-06（万方平台首次上网日期，不代表论文的发表时间）