A probabilistic distance measure for hidden Markov models.

We propose a probabilistic distance measure for measuring the dissimilarity between pairs of hidden Markov models with arbitrary observation densities. The measure is based on the Kullback- Leibler number and is consistent with the reestimation technique for hidden Markov models. Numerical examples which demonstrate the utility of the proposed distance measure are given for hidden Markov models with discrete densities. Effects of various parameter deviations in the Markov models on the resulting distance are discussed.