Zero crossing based feature extraction for a voice command system using neck microphone

Abstract

Automatic speech recognition (ASR) system has been developed over several decades. Although many researchers have found numerous methodologies for improving its performance, the speech processing in noisy environments remains as a difficult problem in this field. One of the solutions is using neck-microphones which are not affected by the environmental noises. However, neck-microphones distort the original voice signals since they only capture the vibrations of vocal tracts. In this context, we consider a method of enhancing features of neck-microphone signals using zero-crossings. Furthermore, we also consider using the concept of gray level co-occurrence matrix (GLCM) which has been usually used for image processing like texture analysis. In this paper, the GLCM is applied to find the suitable zero-crossing features for speech recognition. Through the simulation for speech recognition using the neck-microphone voice command system, we have shown the suggested method provides the better performance than other approaches using conventional speech features.