Speech Recognition for Finite Passwords in Smart Home

DOI

10.2991/ceis-16.2016.72How to use a DOI?

Keywords

finite passwords; VAD; improved DTW; teager energy

Abstract

Dynamic Time Warping (DTW) is the most widely used algorithm for finite passwords or isolated words recognition system. However, the recognition rate using DTW is strongly influenced by the precision of voice activity detection(VAD), and the traditional DTW algorithm has more path to search when compared with the improved DTW algorithm will cost more time. To solve the problem, a novel recognition method for limited passwords using a new VAD algorithm of improved short-time Teager energy with double dynamic thresholds combined with the improved DTW algorithm for limited passwords or isolated words recognition is proposed in this paper. Experiment results show that the proposed method has better noise robustness and recognition rate than the short-time energy and zero-crossing rate with DTW or the improved DTW algorithm. Because of using the improved DTW, it is also real-time.

Copyright

© 2017, 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  - Jun Yang
AU  - Cui-Yun Gao
AU  - Ming Zhang
PY  - 2016/11
DA  - 2016/11
TI  - Speech Recognition for Finite Passwords in Smart Home
BT  - Proceedings of the 2016 International Conference on Computer Engineering and Information Systems
PB  - Atlantis Press
SP  - 359
EP  - 362
SN  - 2352-538X
UR  - https://doi.org/10.2991/ceis-16.2016.72
DO  - 10.2991/ceis-16.2016.72
ID  - Yang2016/11
ER  -