Application of compressed sensing theory in the sampling and reconstruction of speech signals

DOI

10.2991/ifmeita-17.2018.68How to use a DOI?

Keywords

Compressed Sensing; Speech Signals; DCT; Orthogonal Matching Pursuit Algorithm

Abstract

This paper studies the application of compressed sensing theory in speech signal sampling and reconstruction of speech signals. According to the sparsity of speech signals in the discrete cosine transform basis (DCT), we propose a speech compressed sensing (CS) system based on DCT domain which realizes sparse representation of speech signal in DCT domain. Utilizing Gauss random matrix as the measurement matrix and orthogonal matching pursuit algorithm (OMP), the performance of speech signal reconstruction is acquired. The simulation results show that the sparsity of the speech signal is higher in the DCT domain and the OMP algorithm can effectively improves the performance of reconstructed speech signals.

Copyright

© 2018, 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  - Xiwen Tang
AU  - Shilong Wu
AU  - Rui Dong
AU  - Guang Xia
PY  - 2018/02
DA  - 2018/02
TI  - Application of compressed sensing theory in the sampling and reconstruction of speech signals
BT  - Proceedings of the 2nd International Forum on Management, Education and Information Technology Application (IFMEITA 2017)
PB  - Atlantis Press
SP  - 406
EP  - 410
SN  - 2352-5398
UR  - https://doi.org/10.2991/ifmeita-17.2018.68
DO  - 10.2991/ifmeita-17.2018.68
ID  - Tang2018/02
ER  -