Background noise reduction via dual-channel scheme for speech recognition in vehicular environment

An effective dual-channel noise reduction method is proposed and implemented to increase the performance of speech recognition in vehicular environment. While various single channel methods have already been developed and dual-channel methods have been investigated somewhat, their effectiveness in real environments, such as in vehicle, has not yet been formally proven in terms of achieving acceptable performance level. Our aim is to remedy the low performance of the single and dual-channel noise reduction methods. In particular, we propose a dual-channel noise reduction method based on a high-pass filter and front-end processing of the eigen-decomposition method. Representative experiments were conducted with a real multi-channel vehicular corpus and results were compared with each other in various multiple-microphone arrangements. From the analysis and results, we show that the enhanced eigen-decomposition method combined with high-pass filter indeed significantly improves the speech recognition performance under dual-channel environment.