User Evaluation of a Multi-dimensional Statistical Dialogue System
This work addresses data efficiency for developers of spoken dialogue systems, but it is incremental as it demonstrates equivalence rather than improvement.
The authors tackled the challenge of reducing data needs in spoken dialogue systems by introducing a multi-dimensional statistical dialogue manager that leverages domain-independent dimensions, showing through a user study that its performance is equivalent to a one-dimensional baseline trained from scratch.
We present the first complete spoken dialogue system driven by a multi-dimensional statistical dialogue manager. This framework has been shown to substantially reduce data needs by leveraging domain-independent dimensions, such as social obligations or feedback, which (as we show) can be transferred between domains. In this paper, we conduct a user study and show that the performance of a multi-dimensional system, which can be adapted from a source domain, is equivalent to that of a one-dimensional baseline, which can only be trained from scratch.