Speech analysis: Desktop items activation using Dynamic time warping algorithm

Advances in speech technology and computing power have created a surge of interest in the practical application of speech recognition. The main goal of this paper is to facilitate recognition accuracy, with emphasis on acoustic and language modeling. With the Dynamic time warping algorithm, recognition of speech commands and generation of the commands for desktop items activation has been done in this paper. Practical speech recognition also requires the computation to be carried out in real time within the limited resources i.e CPU power and memory size of commonly available computers. In this paper we focus on implementing dynamic time warping algorithm and compare the results with viterbi search algorithm so that speech recognition systems that are insensitive to the constraints in the speech signal and attain human like recognition performance can be identified and can be used for commercial use.