A Speaker Independent Isolated Word Recognition System for Turkish
In this paper, a speaker independent isolated word recognition system based on hidden Markov models (HMM) having a vocabulary of 12 Turkish words is presented. Several LPC representations are tested as feature vectors and the obtained results are compared.
KeywordsFeature Vector Hide Markov Model Speech Recognition Speech Signal Recognition Accuracy
Unable to display preview. Download preview PDF.
- J. R. Deller, Jr., J. G. Proakis, J. H. L. Hansen, Discrete-Time Processing of Speech Signals. MacMillan Publishing Company, New York, 1993Google Scholar