Vietnamese Voice Recognition for Home Automation using MFCC and DTW Techniques

Home automation with voice recognition can achieve a high level of performance in real world environment. However, such performance drops significantly in mismatched noisy conditions. To solve this problem, we propose a improvement method to extract Mel Frequency Cepstral Coefficients (MFCC) that increase the accuracy up to 20% than traditional method. This paper describes an approach of speech recognition for home automation in Vietnamese language by using the improvement MFCC and Dynamic Time Warping (DTW).