A system for dynamically adjusting a voice recognition system is provided herein. The system employs a methodology for determining a display associated with a disambiguation window, with the determination employing a detection system scoring to display disambiguation based on the vocal clarity of the speaker and other environmental factors that may affect detection.