DSP Implementation of Voice Recognition Using Dynamic Time Warping Algorithm

Abstract

Voice recognition is a process of an automatic system to perceive speech. The paper discusses voice recognition using cepstral analysis and DTW of a set of five words. The software model was designed using DSP block library in Simulink. The developed model provide us with the necessary tools to record, filter, and analyze different voice samples and compare them with the archived sample. The paper concludes with discussion about the implementation of the speech recognition algorithm on a DSP processor. The algorithm was implemented on the target DSP via embedded target for TI DSP toolbox and real time workshop (RTW).