Skip to main content

Table 3 Mean square error comparison of different a priori SNR estimation methods for different noise types

From: An adaptive a priori SNR estimator for perceptual speech enhancement

Noise type DD MDD TSNR Proposed
Pink 32.18 30.48 32.09 3 0 . 1 6
Factory 48.53 47.36 47.50 4 7 . 2 0
Babble 41.19 40.46 40.93 4 0 . 3 3
  1. Bold values denote the best performance