Hidden Markov model adaptation algorithm using maximum a posteriori nonlinear transformation in noisy environments

The performance of speech recognition system will be significantly deteriorated because of the mismatches between training and testing conditions. This paper addresses the problem and proposes an environment adaptation algorithm to adapt the mean vectors of HMM using nonlinear transformation, which is approximated by piecewise linear regression. The algorithm can reduce the performance deterioration of the speech recognition system caused by the mismatches. Rather than estimating the transformation parameters using maximum likelihood estimation (MLE), we proposed to use maximum a posteriori (MAP) as the estimation criterion. The proposed algorithm, called MAPNT, has been evaluated on a Chinese digit recognition experiment based on continuous density HMM. The test shows that the proposed algorithm is efficient and outgoing other algorithms, such as maximum a posteriori linear regression (MAPLR) algorithm and maximum likelihood linear regression (MLLR) algorithm etc..