Phoneme-level parameterization of speech using an articulatory model

An analysis scheme is presented for estimating the phoneme-level articulatory parameters to obtain best fits to natural speech. The working units of optimization are the parameters of an articulatory model (one vector per phoneme) and vectors of time and speed of transition for each parameter. The output of a text-to-speech system is used to initialize these parameters. A single prototype interpolation function is used to generate parameter transitions. Results demonstrate that synthesis with phoneme-level units can produce speech comparable to that produced by reasonably good frame-by-frame speech coders.