Voice activity detection based on using wavelet packet

This paper, presents a robust voice activity detection (VAD) technique based on wavelet packet. In this technique sub-bands and their amplitudes are represented as the vectors for each sample time in order to find a new feature from the frequency and amplitude changes. On the other hand, the multi-r...

Full description

Saved in:
Bibliographic Details
Published inDigital signal processing Vol. 20; no. 4; pp. 1102 - 1115
Main Authors Eshaghi, Mohadese, Karami Mollaei, M.R.
Format Journal Article
LanguageEnglish
Published Elsevier Inc 01.07.2010
Subjects
Amplitudes
Discrete wavelet packet
Feature extraction
Mathematical analysis
Speech
Speech processing
Transforms
Vectors (mathematics)
Voice
Voice activity detection
Wavelet
Voice activity detection
Discrete wavelet packet
Speech processing
Online AccessGet full text

Cover

More Information
Summary:This paper, presents a robust voice activity detection (VAD) technique based on wavelet packet. In this technique sub-bands and their amplitudes are represented as the vectors for each sample time in order to find a new feature from the frequency and amplitude changes. On the other hand, the multi-resolution analysis property of the wavelet packet transform (WPT), the voiced, unvoiced, and transient components of speech can be distinctly discriminated. Then, a new feature extraction method is implemented based on observations of the angles between vectors. This feature extraction method retains most unvoiced sounds in a voice active frame. Experimental results show that the proposed WT feature parameter can extract the speech activity under poor SNR conditions and that it is also insensitive to variable-level of noise.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:1051-2004
1095-4333
DOI:10.1016/j.dsp.2009.11.008