One of the difficult problems facing teachers of the deaf and speech clinicians is the lack of tools for training prosody (i.e., intonation, stress, intensity, duration, timing and syllable rhythm). Aside from a few audiocassettes (e.g., Allen's audio-tutorial) and some expensive, yet not very interactive hardware (e.g., VisiPitch), there are no materials available for prosody training in schools for the deaf or speech clinics. The Interactive Prosody Trainer is a low-cost, microcomputer-based system for interactively teaching speech prosody. Based on a single digital signal processing chip and an easy-to-use graphical interface, this device does two different, interrelated jobs. On output from the host microcomputer, it synthesizes models of utterances from stored linear prediction coefficients and prosodic control parameters. On input, it extracts the fundamental frequency and intensity of the user's productions, for comparison with the model as presented. The host microcomputer then visually displays the two productions (i.e., the original model and the user's response) with similarities and differences highlighted for the user's information. Appropriately structured presentations can then lead the user through increasingly precise mimicry of more and more complex computer-generated models.
