Distraction Potential of Speech-Based Driver Interfaces

A common assumption concerning speech-based interaction with an invehicleinformation system is that the speech-based interaction does not distract driver,because the driver is not required to take his eyes off the road. This assumption does nottake into consideration the cognitive demand placed on the driver. This cognitive demandmay be highly dependent upon the nature of the interaction and may increase when errorsoccur in the interchange between the driver and the speech-based system. When theautomatic speech recognition system makes an error, the driver must first recognize thatan error has been made, determine how to recover from the error, trace back to theprevious menu, and repeat the command to get the desired result. These additional stepsand the error recovery process may place significant cognitive demands on driver.Understanding how these errors and the recovery process affects driver attention to theroad is a critical design consideration for speech-based interaction with in-vehicleinformation systems. This paper describes an initial experiment to address this issue andprovides a theoretical framework to help identify the requirements of a speech-interfaceneeded to support easy error recovery. Because speech interactions will always besubject to human and system error, understanding how to support the robust interaction iscritical in minimizing driver distractio