A fuzzy information space approach to speech signal non-linear analysis
Wladimir Rodríguez, Horia-Nicolai Teodorescu, Florian Grigoras, Abraham Kandel, Horst Bunke · International Journal of Intelligent Systems · 2000
A new approach for analyzing the similarity of dynamical systems is presented, with applications to speech analysis. This approach is based on a temporal fuzzy set representation of the trajectories of the dynamical system. The similarity between segments of the speech signal is determined via similarity measures of the corresponding temporal fuzzy sets. We present an application of the method to vowel recognition in the samples (amplitude–time) space. © 2000 John Wiley & Sons, Inc.