In this report, we start with a revisit to the statistical formulation of the automatic speech recognition (ASR) problem, and identify the factors which might influence the performance of the conventional plug-in MAP decision rule for ASR. We summarize our recent research efforts on a class of robust speech recognition problems in which mismatches between training and testing conditions exist but an accurate knowledge of the mismatch mechanism is unknown. The only available information is the test data along with a set of pre-trained speech models and the decision parameters. We focus on two types of Bayesian techniques, namely online Ba yesian adaptation of hidden Markov model parameters and the Bayesian predictive classification approach. We conclude the report with a brief mention of our ongoing research efforts towards a robust and intelligent spoken dialogue system.