Comparative Study of Masking and Mapping Based on Hierarchical Extreme Learning Machine for Speech Enhancement
Ryandhimas Edo Zezario, Join W. C. Sigalingging, Tassadaq Hussain, Jia‐Ching Wang, Yu Tsao · 2019
In this study, we compare the speech enhancement performance based on hierarchical extreme learning machine (HELM) with two distinct strategies: masking and mapping. Experimental results of the perceptual evaluation of speech quality (PESQ) show that for both of limited and sufficient amounts of training data, mapping-based HELM tends to be more effective to improve the performance of speech enhancement.