A continuous speech recognition system using finite state network and Viterbi beam search for the automatic interpretation

This paper describes a Korean continuous speech recognition system using phone based semi-continuous hidden Markov model (SCHMM) method for automatic interpretation. The task domain is hotel reservation. The system (composed of speech recognition, machine translation and speech synthesis) has the following three features. First, an embedded bootstrapping training method is used that enables us to train each phone model without the need for a phoneme segmentation database. Second, a hybrid estimation method which is composed of the forward-backward algorithm and the Viterbi algorithm is proposed for the HMM parameter estimation. Third, a between-word modeling technique is used at the function word boundaries. The recognition results in speaker independent experiments are as follows. In the case of Version 1, the continuous speech recognition result is 89.1% and in Version 2, the result is 97.6%