Application of Bat Algorithm in dual channel speech enhancement

A new speech enhancement approach is proposed to dual channel enhancement systems based on a recently developed meta-heuristic algorithm called Bat Algorithm (BA). Bat algorithm is a bio inspired algorithm which is based on echolocation behavior of micro bats. BA uses frequency technique to increase the diversity in the search space and it also uses automatic zooming effect to provide a dynamic balance between exploration and exploitation in the search process. This paper aims to present the Bat algorithm as a very promising optimization technique than Standard PSO algorithm, in the context of speech enhancement. Simulation results prove that BA algorithm excels than standard PSO algorithm in the application of speech enhancement from both the aspects of improved quality and intelligibility of enhanced speech.