ISOLATED WORD RECOGNITION USING MARKOV-CHAIN MODELS

The paper describes how Markov chains may be applied to speech recogni tion. In this application, a spectral vector is modeled by a state of the Markov chain, and an utterance is represented by a sequence of sta tes. The Markov chain model (MCM) offers a substantial reduction in co mputation, but at the expense of a significant increase in memory requ irement when compared to the hidden Markov model (HMM). Experiments on isolated word recognition show that the MCM achieved results that are comparable to those of the HMM's tested for comparison.