Modeling acoustic transitions in speech by state-interpolation hidden Markov models

We present a new type of HMM for vowel-to-consonant (VC) and consonant-to-vowel (CV) transitions based on the locus theory of speech perception. The parameters of the model can he trained automatically using the Baum-Welch algorithm and the training procedure does not require that instances of all possible CV and VC pairs be present. When incorporated into an isolated word recognizer with a 75 000 word vocabulary we find that it leads to a modest improvement in recognition rates.