Bayesian adaptive learning of the parameters of hidden Markov model for speech recognition

In this paper a theoretical framework for Bayesian adaptive learning of discrete HMM and semi continuous one with Gaussian mixture state observation densities is presented Corre sponding to the well known Baum Welch and segmental k means algorithms respectively for HMM training formulations of MAP maximum a posteriori and segmental MAP estima tion of HMM parameters are developed Furthermore a computationally e cient method of the segmental quasi Bayes estimation for semi continuous HMM is also presented The important issue of prior density estimation is discussed and a simpli ed method of moment estimate is given The method proposed in this paper will be applicable to some prob lems in HMM training for speech recognition such as sequential or batch training model adaptation and parameter smoothing etc