A possibility for improvement of different speech processing systems

In this paper we investigate the possibility to improve performances of different speech processing systems by focusing the analysis on the speech signal segments named sub-phonemic acoustic targets. Presented is theoretical background necessary to realize this analysis. It includes appropriate symbolic description of the speech signal at word and sub-word level and formulation of IFC-guided analysis of the speech signal to localize sub-phonemic acoustic targets. New experimental evidence ‘covering’ several phonemic speech segments belonging to the same word confirmed the possibility to reduce speech signal processing only on the portions corresponding to sub-phonemic acoustic targets. Possible new research projects are suggested.