Importance-Driven Turn-Bidding for Spoken Dialogue Systems
Selfridge, Ethan and Heeman, Peter

Article Structure

Abstract

Current tum-taking approaches for spoken dialogue systems rely on the speaker releasing the turn before the other can take it.

Introduction

As spoken dialogue systems are designed to perform ever more elaborate tasks, the need for mixed-initiative interaction necessarily grows.

Current Turn-Taking Approaches

Current dialogue systems focus on the release-turn as the most important aspect of turn-taking, in which a listener will only take the turn after the speaker has released it.

Importance-Driven Turn-Bidding (IDTB)

We introduce the IDTB model to overcome the deficiencies of current approaches.

Information State Update and Reinforcement Learning

We build our dialogue system using the Information State Update approach (Larsson and Traum, 2000) and use Reinforcement Learning for action selection (Sutton and Barto, 1998).

Domain and Turn-Taking Models

In this section, we show how the IDTB approach can be implemented for a collaborative slot filling domain.

S: i. avail slot fillers i. burger have bl

S: bye

Topics