Abstract: “In this paper, we analysed the performance analysis for speech enhancement using spectral subtraction, affine projection algorithms and classical adaptive filters. As the enhancement of speech signals is of very important in many applications like speech recognition, hearing aids, forensic applications and telephone conversations etc. The performance of the speech recognition also reduces if the speech signal is corrupted by noise. To remove the noise present in the speech signal, the adaptive filters shown the good improvement in increasing the Signal to Noise Ratio (SNR) values. The matlab simulations are performed using NOIZEUS speech corpus for different SNR values using Spectral Subtraction(SS), Least Mean Square (LMS), Normalized Least Mean Square (NLMS), Recursive Least Squares (RLS), Affine Projection Algorithms (APA). From the results it is observed that RLS algorithm has shown good improvement in speech enhancement when compared to the other methods.”

Keywords: NLMS, Affine Projection, RLS,Speech Enhancement