A Novel Algorithm to Robust Speech Endpoint Detection in Noisy Environments

摘要：

Accurate endpoint detection is crucial for good speech recognition accuracy. A new algorithm based on C0 complexity measure is proposed in this paper. In the comparison of the new algorithm to the other traditional methods such as energy, ZCR, spectral-entropy etc., for the continuous speech and isolate word speech, the C0 complexity feature proved effective. The proposed algorithm is shown to be well suited for the detection of speech endpoint and is very robust for different types of noise, especially for low SNR.

作者: Yi LI Yingle FAN

作者单位: Hangzhou Dianzi University, China

会议类型: 国际会议

会议名称: 2nd IEEE Conference on Industrial Electronics and Applications(ICIEA 2007)(第二届IEEE工业电子与应用国际会议)

会议地点: 哈尔滨

会议语种:英文

在线出版日期: 2007-05-23（万方平台首次上网日期，不代表论文的发表时间）