Fast Wavelet-based Pitch Period Detector for Speech Signals

DOI

10.2991/ceis-16.2016.101How to use a DOI?

Keywords

pitch period; haar wavelet function

Abstract

A fast algorithm which can extract pitch period information from speech is described. The solution is developed by processing speech signals using Haar wavelet function. We have shown mathematically that the pitch period of a speech signals can be computed using global minimum of the total energy level of the Haar processed signal. The accuracy of the proposed method is examined by observing the differences in pitch period obtained between the proposed method and the ETSI pitch estimation algorithm. By reducing the computational complexity, proposed pitch detector is 8 times faster than the conventional ETSI pitch estimation algorithm. Besides, the proposed method has been implemented in a energy-based harmonic-features speech recognition system, recognition results have been compared and discussed.

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  - Y.H. Goh
AU  - Y.H Ko
AU  - Y.K Lee
AU  - Y.L Goh
PY  - 2016/11
DA  - 2016/11
TI  - Fast Wavelet-based Pitch Period Detector for Speech Signals
BT  - Proceedings of the 2016 International Conference on Computer Engineering and Information Systems
PB  - Atlantis Press
SP  - 494
EP  - 497
SN  - 2352-538X
UR  - https://doi.org/10.2991/ceis-16.2016.101
DO  - 10.2991/ceis-16.2016.101
ID  - Goh2016/11
ER  -