Computer speech recognition provides the enabling technology for the use of automated performance measurement and instructor support features to allow "instructorless" training for those tasks which are primarily speech in nature. The design guides take the approach that automated, "instructorless" training is achievable through the use of computer software models of the instructor and the task. Human factors design guidelines are provided for the integration of speech technology with the software models. The Navy has built two prototype training systems using voice technology to capture student behavior, evaluated one of them, and is in the process of evaluating another. Further, the training implications for airborne applications of voice technology have been developed. The emergence of voice technology as one solution to the manpower shortage has provided justification for these efforts. This paper describes how voice technology can make the transition from R&D to application.