Perceptual Quality Dimensions of Spoken Dialogue Systems: a Review and New Experimental Results

Spoken dialogue systems, i.e. computer systems which enable a more-or-less “natural” human-machine interaction via spoken language, become increasingly popular for a number of applications. When interacting with such a system, the human user may differentiate several perceptual dimensions which contribute to the overall quality of the service. This paper reviews multidimensional analyses which have been carried out in the past to identify these dimensions. The results are compared to new experimental data which have been obtained with the INSPIRE system for controlling domestic devices. This analysis reveals five categories of quality aspects which can be assumed to have general significance for the perceived quality, across a larger number of systems and services.