Using Affordances to shape the interaction in a hybrid spoken dialogue system


Affordances manifest possibilities of interaction with a spoken dialog system. For example, the act of asking a question affords to the recipient the possibility of answering. In the system we present, the observable act of maneuvering affords the possibility of controlling a motion. Our system thus uses the affordance principle to shape the interaction: to trigger the usage of instructions that are easy to understand and process, the system gives immediate visual feedback to afford user commands that can then be reacted upon. This tightening of the interaction loop requires an incremental processing paradigm to allow fast reactions and to be able to alter ongoing system actions. Our system is a hybrid of incremental and non-incremental processing components, combining conventional, state graph-based processing, which has the advantage of widely available toolkits and well-understood dialog management, with incremental dialog processing which allows for the tight feedback loop that provides for quick reactions. We tested our approach in a small user study and found that users used simpler and setting-independent commands more often and were more efficient when faced with the affordance-based version of our system.

Year: 2013
In session: Dialogsysteme
Pages: 12 to 19