A Sequence-to-Sequence Model for User Simulation in Spoken Dialogue Systems

TL;DR

This paper presents a neural sequence-to-sequence model for user simulation in spoken dialogue systems, effectively capturing dialogue history and user intentions, and outperforming traditional models on benchmark datasets.

Contribution

Introduces a data-driven encoder-decoder RNN model for user simulation that handles dialogue history and multiple user intentions, improving over prior methods.

Findings

Outperforms agenda-based and n-gram simulators on DSTC2 dataset

Models user behavior with finer granularity using original action space

Achieves higher F-score in user simulation tasks

Abstract

User simulation is essential for generating enough data to train a statistical spoken dialogue system. Previous models for user simulation suffer from several drawbacks, such as the inability to take dialogue history into account, the need of rigid structure to ensure coherent user behaviour, heavy dependence on a specific domain, the inability to output several user intentions during one dialogue turn, or the requirement of a summarized action space for tractability. This paper introduces a data-driven user simulator based on an encoder-decoder recurrent neural network. The model takes as input a sequence of dialogue contexts and outputs a sequence of dialogue acts corresponding to user intentions. The dialogue contexts include information about the machine acts and the status of the user goal. We show on the Dialogue State Tracking Challenge 2 (DSTC2) dataset that the sequence-to-sequence model outperforms an agenda-based simulator and an n-gram simulator, according to F-score. Furthermore, we show how this model can be used on the original action space and thereby models user behaviour with finer granularity.