Incremental LSTM-based Dialog State Tracker

TL;DR

This paper introduces an incremental dialog state tracker using LSTM networks that leverages ASR hypotheses and novel techniques to improve performance in spoken dialog systems.

Contribution

It presents a new LSTM-based dialog state tracker with innovative features like confidence scores and data augmentation, achieving near state-of-the-art results.

Findings

Performance close to state-of-the-art

Effective use of ASR confidence scores

Benefits of including transcriptions in training

Abstract

A dialog state tracker is an important component in modern spoken dialog systems. We present an incremental dialog state tracker, based on LSTM networks. It directly uses automatic speech recognition hypotheses to track the state. We also present the key non-standard aspects of the model that bring its performance close to the state-of-the-art and experimentally analyze their contribution: including the ASR confidence scores, abstracting scarcely represented values, including transcriptions in the training data, and model averaging.