Dynamic Programming Algorithm Optimization for Spoken Word Recognition



This paper reports on an optimum dynamic programming (DP) based time-normalization algorithm for spoken word recognition. First, a general principle of time-normalization i sgiven using time- warping function. Then, two time-normalized distance definitions, d e d symmetric and asymmetric forms, are derived from the principle. These two forms are compared with each other through theoretical discussions and experimental studies. The symmetric form algorithm superiority is established. A new technique, called slope constraint, i s successfully introduced, in which the warping function slope is restricted so as t o improve discrimination between words in different categories. The effective slope constraint characteristic is qualitatively analyzed, and the optimum slope constraint condition is determined through experiments. The optimized algorithm i s then extensively subjected to experimentat comparison with various DP-algorithms, previously applied t o spoken word recognition by different research groups. The experi- ment shows that the present algorithm gives n o more than about two- thirds errors, even compared t o the best conventional algorithm.

Click here for free

download this paper


CSE PROJECTS

FREE IEEE PAPER AND PROJECTS

FREE IEEE PAPER