Analysis and Comparison of Evolutionary Algorithms applied to Adaptive Noise Cancellation for Speech Signal

In this paper, an improved method based on evolutionary algorithm for speech signal denoising is proposed. In this approach, the stochastic global optimization techniques such as Artificial Bee Colony(ABC), Cuckoo Search (CS)algorithm, and Particle Swarm Optimization (PSO) technique are exploited for learning the parameters of adaptive filtering function required for optimum performance. It was found that the ABC algorithm and Cuckoo Search algorithm based speech denoising approach give better performance in terms of signal-to-noise ratio (SNR) as compared to PSO based speech denoising approach. The quantitative (SNR, MSE and Maximum Error (ME)) and visual (Denoised speech signals) results show superiority of the proposed technique over the conventional and state -of-art speech signal denoising techniques. All proposed methods have been simulated in Matlab, and design results are illustrated clearly to show the superiority of the proposed method. Keywords— Adaptive filters, Adaptive Algorithms, Artificial Bee Colony, Cuckoo search, Particle Swarm Optimization Algorithm, Speech Enhancement.