Improved Particle Swarm Optimization for Dual-Channel Speech Enhancement

Author

Asl, Laleh Badri ; Nezhad, Vahid Majid

Author_Institution

Fac. of Electr. & Comput. Eng., Univ. of Tabriz, Tabriz, Iran

fYear

2010

fDate

9-10 Feb. 2010

Firstpage

13

Lastpage

17

Abstract

This paper, proposes an improved particle swarm optimization algorithm for speech enhancement. In the proposed algorithm, the population is divided into two subgroups. One of the subgroups searches the space globally, whereas the other one explores the problem space locally. The proposed algorithm surpasses the standard PSO, by stimulating the inactive particles and local search around the global best. Experimental results indicate that improved particle swarm optimization (IPSO) outperforms the standard particle swarm optimization (SPSO), and gradient-based NLMS algorithm in dual-channel speech enhancement applications.

Keywords

gradient methods; particle swarm optimisation; speech enhancement; dual-channel speech enhancement; gradient-based NLMS algorithm; improved particle swarm optimization; population division; standard particle swarm optimization; Application software; Filtering; Kalman filters; Least squares approximation; Particle swarm optimization; Signal processing; Space exploration; Speech enhancement; Stochastic processes; Wiener filter; Improved Particle Swarm Optimization (IPSO); Speech Enhancement; Standard Particle Swarm Optimization (SPSO);

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Acquisition and Processing, 2010. ICSAP '10. International Conference on

Conference_Location

Bangalore

Print_ISBN

978-1-4244-5724-3

Electronic_ISBN

978-1-4244-5725-0

Type

conf

DOI

10.1109/ICSAP.2010.30

Filename

5432631