Incorporation of State-Level Variable Stime-Varying Property into the HMM

In HMM-based pattern recognition, the structure of the HMM is often predetermined according to some prior knowledge. In the recognition process, we usually make our judgment based on the maximum likelihood of the HMM, without considering the time-varying property of state-level variables, which unfortunately may lead to incorrect results. In this paper, we analyze the property of state-level variables in the HMM and show it is possible to significantly enhance the performance of speech recognition systems when using the state-level variable time-varying property. We propose to make use of the distribution of the number of intersecting points (NIPs) of state-level variable trajectories in the recognition process, which achieve 2.7 percent correct improvement to a phoneme classification task on TIMIT speech corpus. We also compare the proposed method and state duration model and draw some empirical conclusions