Improved parameter tying for efficient acoustic model evaluation in large vocabulary continuous speech recognition

In an HMM based large vocabulary continuous speech recognition system, the evaluation of context dependent acoustic models is very time consuming. In Semi-Continuous HMMs, a state is modelled as a mixture of elementary generally gaussian probability density functions. Observation probability calculations of these states can be made faster by reducing the size of the mixture of gaussians used to model them. In this paper, we propose di erent criteria to decide which gaussians should remain in the mixture for a state, and which ones can be removed. The performance of the criteria is compared on context dependent tied state models using the WSJ recognition task. Our novel criterion, which decides to remove a gaussian in a state if it is based on too few acoustic data, outperforms the other described criteria.