Hybrid training method for tied mixture density hidden Markov models using learning vector quantization and Viterbi estimation

In this work the output density functions of hidden Markov models (HMMs) are phoneme-wise tied mixture Gaussians. For training these tied mixture density HMMs, modified versions of the Viterbi training and learning vector quantisation (LVQ) based corrective tuning are described. The initialization of the mean vectors of the mixture Gaussians is performed by first composing small self-organising maps representing each phoneme and then combining them to a single large codebook to be trained by LVQ. The experiments on the proposed training methods are accomplished using a speech recognition system for Finnish phoneme sequences. Comparing to the corresponding continuous density and semi-continuous HMMs regarding the number of parameters, the recognition time and the average error rate, the performance of the phoneme-wise tied mixture HMMs is superior