Synthesized stereo-based stochastic mapping with data selection for robust speech recognition

Abstract

In this paper, we present a synthesized stereo-based stochastic mapping approach for robust speech recognition. We extend the traditional stereo-based stochastic mapping (SSM) in two main aspects. First, the constraint of stereo-data, which is not practical in real applications, is relaxed by using HMM-based speech synthesis. Then we make feature mapping more focused on those incorrectly recognized samples via a data selection strategy. Experimental results on Aurora3 databases show that our approach can achieve consistently significant improvements of recognition performance in the well-matched (WM) condition among four different European languages.