Development and Evaluation of Automotive Speech Interfaces: Useful Information from the Human Factors and the Related Literature

Abstract

Drivers often use infotainment systems in motor vehicles, such as systems for navigation, music, and phones. However, operating visual-manual interfaces for these systems can distract drivers. Speech interfaces may be less distracting. To help designing easyto-use speech interfaces, this paper identifies key speech interfaces (e.g., CHAT, Linguatronic, SYNC, Siri, and Google Voice), their features, and what was learned from evaluating them and other systems. Also included is information on key technical standards (e.g., ISO 9921, ITU P.800) and relevant design guidelines. This paper also describes relevant design and evaluation methods (e.g., Wizard of Oz) and how to make driving studies replicable (e.g., by referencing SAE J2944). Throughout the paper, there is discussion of linguistic terms (e.g., turn-taking) and principles (e.g., Grice's Conversational Maxims) that provide a basis for describing userdevice interactions and errors in evaluations.

Using speech, GUIs and buttons in police vehicles

“…Furthermore, while a number of researchers have found evidence that speech-based HMIs can provide for better driving performance [6], it is clear that speech input/output is not always a good match to in-vehicle tasks, especially if for tasks that are simple, such as flipping a switch. On the other hand, drivers are likely to be open to using speech interfaces for complex tasks, which are exactly the tasks where good design can improve driving safety.…”

“…However, as Lo and Green pointed out, there is almost no field data on who uses speech interfaces, for what, and how often [6]. Moreover, to the best of our knowledge, there is almost no field data on how police officers use in-vehicle interfaces.…”

Considering Self-Report in the Interpretation of Objective Performance Data in the Comparison of HMI Systems

“…The potential benefits of voice-based interactions relative to visual-manual interaction have been relatively well demonstrated in simulated or prototype implementations (e.g., see reviews in Andre & Wickens, 1995;Lo & Green, 2013;Reimer, Mehler, Dobres & Coughlin, 2013). However, relatively few studies are available on the extent to which these benefits are observed with actual production systems, and even fewer have examined production systems under onroad conditions (see summaries in Mehler et al, 2016;Reimer et al, 2013).…”

A Secondary Assessment of the Impact of Voice Interface Turn Delays on Driver Attention and Arousal in Field Conditions

“…Other studies focus more on voice recognition accuracy, task time, and user preference (e.g., Walker, Kamm, & Litman, 2000;Hajdinjak & Mihelic 2006). These evaluation methods often face criticism for neglecting the demand that various systems place upon the user (Hajdinjak & Mihelic 2006;Lo & Green, 2013). Holistic assessments of the demands of voice systems need to consider all potential factors to support the development of highly optimal interfaces that minimize the impact on attention and driving behavior .…”