Project Aphatar : An Experiment and Implementation

Abstract

Aphasia is an acquired communication disorder that affects the ability of a person to speak and understand spoken language. The purpose of the Aphatar project is to create a virtual clinician that will help suffers of aphasia improve their speech in common scenarios. The project will gauge the interaction and quality of this virtual clinician against those of a real clinician. Aphatar will be created using three major systems: (1) KINECT for audio and visual recording, audio input, and future work in reading the client's emotional state using the KINECT 3D Camera system, (2) The Olympus Speech Recognition System, provided by Carnegie-Mellon University which will accept the audio input of the user and translate it from speech to text then provide spoken feedback to the user, and (3) the Avatar display system, which will provide the graphical interface for the former, allowing the user to see the avatar and interact with it.