On Some Applications of Fuzzy Algorithm in Man-Machine Communication Research

Abstract

Present paper ennuuciates some practical applications of fuzzy set theory in problems of man-machine communication. Problems are recognition of vowel speech sounds and identification of speakers from spoken words. Data used are derived from acoustic-phonetic and spectrographs analysis of large number of Hindi and Telugu (two of the major Indian Languages) speech sounds. It is explained that Fuzzy set theory provides a suitable algorithm which is substantively different from the conventional quantitative methods of system analysis yet presents an approximate but effective means of describing the behaviour of systems which are too ill-defined for precise mathematical analysis. In this paper, two decision algorithmic methods developed from fuzzy set theory and applied for machine recognition of vowels and identification of speakers with Telugu and Hindi speech sounds are presented along with the results of experiments.

Indexing Terms:

• Fuzzy Set
• Format Frequencies
• Membership Function
• Property Set
• Heighting Co-efficient
• Similarity Vector
• Pattern Recognition
• Recognition Score