Noise robust speech recognition using parallel model compensation and voice activity detection methods

Abstract

The main purpose of this study is to increase the performance of a speech recognition system under noisy environments. In this study Voice Activity Detection (VAD) methods is used for estimating the noise model, and Parallel Model Compensation (PMC) is used for estimating the noisy speech model using the clean speech model and noise model which is estimated using a VAD method. Performances of the baseline and four well known VAD methods have been compared for noisy speech recognition. In addition to this, a new VAD method is proposed to estimate parameters of the noise model. The proposed VAD method's speech recognition performance is better than the most of the well-known VAD methods despite less computational requirement of the proposed VAD method compared to these well-known VAD methods.