Stochastic Modelling of Sentence Semantics in Speech Recognition

A stochastic approach to spoken sentence recognition is proposed for the purpose of an automatic voice-based dialogue system. Three main tasks are distinguished: word recognition, word chain filtering and sentence recognition. The first task is solved by typical acoustic processing followed by phonetic word recognition with the use of Hidden Markov Models (HMM) and Viterbi search. For the second solution an N-gram model of natural language is applied and a token-passing search is designed for the filtering of important word chains. The third task is solved due to a semantic HMM of sentences. The final sentence is recognized and a meaning is assigned to its elements with respect to given application domain. A particular spoken sentence recognition system has been implemented for train connection queries.