The objective is the exploratory development, miniaturization, and demonstration of a universal and advanced technology voice recognition and synthesis capability suitable for a tactical environment. TERIs active participation in state-of-the-art voice processing systems presents a unique opportunity for the preliminary development of a microchip implementation based upon TERIs experience with Phase I development of a miniaturized architecture, and knowledge of ATCCS and Battlefield Functional Area applications. TERIs candidate architecture demonstrates the feasibility of a wide range of Army personnel verbally communicating with Army C2 applications as a replacement for or adjunct to more traditional 1/0 devices (i .e., keyboard, trackball, display, etc.). Common military language input is recognizable and acted upon by the C2 application then the TERI developed interface drivers interpret commands using an ATCCS universal multi-layered API. Computer generated voice output demonstrates the effective combination of candidate methods, including: unlimited text-to-speech synthesis, a predefined unique pronunciation library, and a high quality digital reproduction capability. Use of sub-micron process technology is proposed for the implementation of common processing module silica that is central to the miniaturization of the voice interactive functions and has DoD wide spin-off application to ASIC and MCM chip packaging and manufacture.
Benefits: A miniaturized voice interaction chip can be used for voice control of a system when an operator's hands and eyes are free for other critical tasks. This improves combat effectiveness and chance for survival. Quality voice output from the system will make a voice processing system accessible to the blind, having all interaction through sljeaking or hearing. Commercial applications include voice-controlled typewriters, security systems, phone systems, safety devices, robotics, inventory, and general replacement of keypads for computer interfaces.
Keywords: Multi Chip Module Synthesis Processing Speech Control Recognition Command Voice