Overview

The GT-Bionics lab is developing a Multimodal Speech Capture System (MSCS) to assist speech-language pathologists (SLPs) in their therapy sessions for many applications, including identifying and correcting errant speech, correct pronunciation and/or accent of second-language learners and native speakers that hold positions in which verbal communication is crucial.

The MSCS can also be used for speech recognition by enabling people, with limited abilities to produce voice, to communicate through text and/or computer-generated voice.

This versatility of applications comes from its capability of recording three speech modalities: tongue motion, lips gesture and voice.