Speech Endpoint Detection Based on EMD and Higher Order Statistics in Noisy Environments

DOI

10.2991/icimm-15.2015.199How to use a DOI?

Keywords

Endpoint Detection; EMD; Higher Order Statistics; Noisy Environments

Abstract

Accurate endpoint detection is crucial for speech recognition accuracy. This paper presents a new technique for speech endpoint detection in a noisy environment based on the empirical mode decomposition (EMD) algorithm and higher order statistics. With the EMD, the noise speech signals can be decomposed into a sum of the band-limited function called intrinsic mode functions (IMFs), which is a zero-mean AM-FM component. Then higher order statistics of the IMF components can be used to extract the desired feature for endpoint detection. In order to show the effectiveness of the proposed method, we present examples showing that the new measure is more effective than traditional measures. The experimental results show that the performance of the proposed algorithm is noticeable in the real speech signal tests with different SNR.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Dexiang Zhang
AU  - Jiaxing Li
AU  - Zihong Chen
PY  - 2015/07
DA  - 2015/07
TI  - Speech Endpoint Detection Based on EMD and Higher Order Statistics in Noisy Environments
BT  - Proceedings of the 5th International Conference on Information Engineering for Mechanics and Materials
PB  - Atlantis Press
SP  - 1101
EP  - 1104
SN  - 2352-5401
UR  - https://doi.org/10.2991/icimm-15.2015.199
DO  - 10.2991/icimm-15.2015.199
ID  - Zhang2015/07
ER  -