Optimization of HMM by The Tabu Search Algorithm

In this paper, the tabu search algorithm is employed to train Hidden Markov Model (HMM) to search out the optimal parameter structure of HMM for automatic speech recognition. The proposed TS-HMM training provided a mechanism that allows the searching process to escape from the local optimum and obtain a near global optimum. Experimental results show that the TS-HMM training has a higher probability to find the optimal model parameters than the traditional algorithms.