INFORMATICA International Journal Main Page Editorial Board Abstracting/Indexing Instructions to Authors Subscription Information Contents Author Index Papers in Production		INFORMATICA, 2004, Vol. 15, No. 4, 465-474 © Institute of Mathematics and Informatics, ISSN 0868-4952 Development of HMM/Neural Network-Based Medium-Vocabulary Isolated-Word Lithuanian Speech Recognition System Mark FILIPOVIC, Antanas LIPEIKA Recognition Processes Department, Institute of Mathematics and Informatics Gostauto 12-204, LT-01108 Vilnius, Lithuania E-mail: markas@mch.mii.lt, lipeika@ktl.mii.lt Abstract The development of Lithuanian HMM/ANN speech recognition system, which combines artificial neural networks (ANNs) and hidden Markov models (HMMs), is described in this paper. A hybrid HMM/ANN architecture was applied in the system. In this architecture, a fully connected three-layer neural network (a multi-layer perceptron) is trained by conventional stochastic back-propagation algorithm to estimate the probability of 115 context-independent phonetic categories and during recognition it is used as a state output probability estimator. The hybrid HMM/ANN speech recognition system based on Mel Frequency Cepstral Coefficients (MFCC) was developed using CSLU Toolkit. The system was tested on the VDU isolated-word Lithuanian speech corpus and evaluated on a speaker-independent \sim750 distinct isolated-word recognition task. The word recognition accuracy obtained was about 86.7%. Keywords: speech recognition, artificial neural networks, hidden Markov models To preview Lithuanian abstract see full article text To preview full article text in PDF format click here You could obtain free Acrobat Reader from Adobe
Top		Copyright © INFORMATICA, Vilnius University Institute of Mathematics and Informatics, 2010