Human-Automation Speech Interface

In applications involving human-automation interactions, a two-way speech interface can be an effective and efficient mode to supplement/complement other more common interface designs involving graphical or haptic modes. For defense-related applications, the speech processing system needs to be standalone without dependence on a bank of powerful servers. The primary Phase II activity for this AFWERX SBIR open topic will be to develop, install, integrate and demonstrate a prototype system for an Air Force unit to be designated by the Air Force Life Cycle Management Center (AFLCMC) Rapid Sustainment Office (RSO). The speech interface technology developed by Optimal Synthesis Inc. (OSI) has gone through several SBIR developments supported by the Navy, DARPA, and NASA, with commercial products deployed under SBIR Phase III awards. The OSI team has successfully applied it to air traffic control (ATC) interaction with UAVs, cockpit interaction between human pilot and flight-control automation, NASA ATC research in replacing human pseudo-pilots with speech agent, and Navy antisubmarine warfare training system. OSI’s speech interface technology has been commercialized as a speech agent under NASA SBIR Phase III awards. The envisioned Air Force product will be a tailorable speech interface for Augmented-Reality/Virtual-Reality (AR/VR)-based maintenance training for aircraft and weapon system platforms.