Fuzzy Logic Networks for Speech Recognition

Abstract

This paper proposes a massive fuzzy logic network which can be considered as a novel model of pattern classification network. Our approach introduces fuzzy logic circuits fulfilling the function of a binary classifier at first, which are connected into fuzzy logic networks with fuzzy flip-flop circuits as memories. Genetic programming is used as a circuit designing method. In order to establish design methodology, experiments aimed at testing the suitability of fuzzy logic operation sets, fitness functions and parameters of genetic algorithm were carried out. From trained circuits a hierarchical layered structure is built, where single layers consisting of given circuits are contextually dependent. Experiments with fuzzy logic circuits and fuzzy flip-flop network show some valuable results especially in the task of audio and visual speech
recognition.