SOAR TECHNOLOGY, LLC — Department of Defense STTR Phase I: N17A-T010

Description

Automatic Speech Recognition (ASR) is the capability where spoken words, commands or phrases are ingested and interpreted by a software system and used to elicit activity or response. ASR capabilities range from dictation, which converts speech into text; to systems that interpret and respond to speech in proper context. In a vast majority of military domains, substantive verbal communication is a critical component of the warfighters primary responsibilities.<br><br>ASR is yet to become a widely-used feature of training systems despite its potential utility. This phenomenon has persisted despite the improvement of spee infeasible for content developers to modify or extend the speech recognition capability to support specific phraseology and terminology. Also, users cannot rehearse best practices when using ASR independent of a full-on exercise. Finally, it is difficult for students to distinguish between features of the ASR system and the connected training system. Extending internally-developed technologies and contemporary speech recognition engine features; we propose Soliloquy: a suite of customization tools designed to empower end-users of simulation systems to employ, customize and practice ASR to better support training needs.