Dual-channel speech enhancement using normalized fractional least-mean-squares algorithm

Summary from only given. In this paper, we propose a novel method, called normalized fractional least-mean-squares (NFLMS) for dual-channel speech enhancement. Here, we use a modified least-mean-squares algorithm known as fractional LMS, by incorporating the fractional term in weight adaptation equation of the standard least-mean-squares algorithm. Normalization provides an interesting way to improve standard least-meansquares performance. In order to promote the performance of the fractional LMS algorithm, we propose normalized version of fractional least-mean squares. Experimental results shows that the proposed method has better performance in dual-channel speech enhancement in a sense of mean-squares error (MSE) and speech quality improvement than the standard LMS, Normalized LMS, and Fractional LMS algorithms.