Title :
Speaker independent voice recognition with a fuzzy neural network
Author :
Nava, Patricia A. ; Taylor, Javin M.
Author_Institution :
Dept. of Electr. & Comput. Eng., Texas Univ., El Paso, TX, USA
Abstract :
Speaker-independent voice recognition is a large and difficult classification problem. A neuro-fuzzy classifier, NFC, that yields excellent classification accuracy, is presented. The NFC is based on the multi-layer perceptron model and incorporates fuzzy theory into its operation. NFC employs the backpropagation with momentum learning rule, with use of cut-sets and interval mathematics to accommodate fuzzy values. Included in the NFC definition are provisions for fuzzification of input, as well as defuzzification of the output. This scheme incorporates the strengths of neural networks and fuzzy systems, thus resulting in more accurate classification as compared to other neural and neuro-fuzzy systems. According to experimental results, the NFC shows better results than several existing methods
Keywords :
backpropagation; fuzzy neural nets; multilayer perceptrons; pattern classification; speech recognition; backpropagation with momentum learning rule; classification accuracy; cut-sets; fuzzy neural network; fuzzy theory; interval mathematics; multi-layer perceptron model; neuro-fuzzy classifier; speaker independent voice recognition; Backpropagation; Computer networks; Fuzzy neural networks; Fuzzy set theory; Fuzzy systems; Multilayer perceptrons; Neural networks; Neurons; Speech recognition; Training data;
Conference_Titel :
Fuzzy Systems, 1996., Proceedings of the Fifth IEEE International Conference on
Conference_Location :
New Orleans, LA
Print_ISBN :
0-7803-3645-3
DOI :
10.1109/FUZZY.1996.552767
