Abstract
Operating in-vehicle information systems (IVISs) while driving is one of the main causes of car accidents. Thus, in order to lessen the risk, many studies have focused on the application of voice recognition, but its accuracy can be highly compromised due to surrounding noises. Thus, some researchers have paid attention to the operation of IVISs using non-touch gestures while driving. In this study, an experiment was conducted to investigate the efficiency and safety of operating IVISs using voices and/or non-touch gestures while driving. For the purpose, the interactions relevant to IVISs were divided into two kinds of operations: the operation itself and its degree (ex. ‘adjust radio volume’ to ‘level 7’). Then, the following four combinations were derived: (1) All Voices, (2) All Gestures, (3) Gesture+Voice (gesture operation with the degree indicated by the voice), and (4) Voice+Gesture (voice operation with the degree indicated by the gesture). The experimental results showed that the ‘All Voices’ group required less operation time and resulted in less errors than the other groups. However, the ‘Voice+Gesture’ group, which could complement the limitation of voice recognition techniques, showed higher subjective satisfaction scores, as well as less operation time and less errors. From the experimental results, this study recommended a multi-modal interaction design based on both voice and non-touch gesture that is likely to be applied in the operation of IVISs.
| Translated title of the contribution | The Effects of Operation Types on Interacting with Voice and Non-Touch Gestures in In-Vehicle Information Systems |
|---|---|
| Original language | Korean |
| Pages (from-to) | 93-102 |
| Number of pages | 10 |
| Journal | 디자인학연구 |
| Volume | 25 |
| Issue number | 2 |
| State | Published - May 2012 |