Fig. 2From: Speech recognition in reverberant and noisy environments employing multiple feature extractors and i-vector speaker adaptationBlock diagram showing various steps for the extraction of multi-taper spectrum estimation-based mel-filterbank (MMFB) features and the conventional mel-filterbank (MFB) features when the number of taper M = 1, λ(p) = 1, and w p (j) is the Hamming taperBack to article page