Bayesian linear regression for Hidden Markov Model based on optimizing variational bounds

Linear regression for Hidden Markov Model (HMM) parameters is widely used for the adaptive training of time series pattern analysis especially for speech processing. This paper realizes a fully Bayesian treatment of linear regression for HMMs by using variational techniques. This paper analytically derives the variational lower bound of the marginalized log-likelihood of the linear regression. By using the variational lower bound as an objective function, we can optimize the model topology and hyper-parameters of the linear regression without controlling them as tuning parameters; thus, we realize linear regression for HMM parameters in a non-parametric Bayes manner. Experiments on large vocabulary continuous speech recognition confirm the generalizability of the proposed approach, especially for small quantities of adaptation data.