Improving Recognition Performance Using Co-articulation Rules on the Phrase Level: A First Approach

This paper describes a first approach to improve recognition performance of our hybrid large vocabulary continuous speech recogniser for Dutch by using co-articulation rules on the phrase level. By applying these rules on the reference transcripts used for training the recogniser and by adding a set of special temporary phones that later on will be mapped on the original phones, more robust models of phones that are typically confused a lot in speech recognition like /v/-/f/ and /s/-/z/, could be trained.