, , , ,

A sea of changes is coming: “So, which one is it?” The effect of alternative incremental architectures in a high-performance game-playing agent

Information Retrieval and Data Mining by conversing with computers is obvious.

Abstract follows:

This paper introduces Eve, a high performance
agent that plays a fast-paced
image matching game in a spoken dialogue
with a human partner. The agent can
be optimized and operated in three different
modes of incremental speech processing
that optionally include incremental
speech recognition, language understanding,
and dialogue policies. We present
our framework for training and evaluating
the agent’s dialogue policies. In a user
study involving 125 human participants,
we evaluate three incremental architectures
against each other and also compare
their performance to human-human game play.
Our study reveals that the most fully
incremental agent achieves game scores
that are comparable to those achieved
in human-human game play, are higher
than those achieved by partially and non incremental
versions, and are accompanied
by improved user perceptions of efficiency,
understanding of speech, and naturalness
of interaction.