A training tool for speech driven human-robot interaction applications

摘要

Following the integration of voice control software with a robotic system at Mississippi State University for use with a tactical team, it became clear a training tool was needed to facilitate a more natural interaction between police officers and a voice-controlled robotic platform. In order to facilitate this more natural interaction, a training tool was designed and developed, which used the same voice recognition software, to prompt the user for input and then score that input, in order to provide them the ability to practice, without being co-located with the robotic system. The design of this system allows this training tool to be used in a variety of settings beyond the police domain for which it was designed. Since the system scores the user's ability to successfully interact with the voice recognition software based on the vocabulary given to the tool, this training tool allows anyone using the Julius voice recognition system the ability to test how successfully they can interact with it and to learn the commands available in the system.