A two-stage domain selection framework for extensible multi-domain spoken dialogue systems

Authors:
Mikio Nakano;Shun Sato;Kazunori Komatani;Kyoko Matsuyama;Kotaro Funakoshi;Hiroshi G. Okuno
Affiliations:
Honda Research Institute Japan, Wako, Saitama, Japan;Tokyo Denki University, Hatoyama, Saitama, Japan;Nagoya University, Nagoya, Aichi, Japan;Kyoto University, Kyoto, Kyoto, Japan;Honda Research Institute Japan, Wako, Saitama, Japan;Kyoto University, Kyoto, Kyoto, Japan
Venue:
SIGDIAL '11 Proceedings of the SIGDIAL 2011 Conference
Year:
2011

Citing 11
Cited 0

Vector-based natural language call routing

Computational Linguistics
Speech technology on trial: Experiences from the August system

Natural Language Engineering
Spoken Dialogue Technology

Spoken Dialogue Technology
Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)

Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)
A markup language for describing interactive humanoid robot presentations

Proceedings of the 12th international conference on Intelligent user interfaces
Example-based dialog modeling for practical multi-domain dialog system

Speech Communication
Style & topic language model adaptation using HMM-LDA

EMNLP '06 Proceedings of the 2006 Conference on Empirical Methods in Natural Language Processing
A framework for building conversational agents based on a multi-expert model

SIGdial '08 Proceedings of the 9th SIGdial Workshop on Discourse and Dialogue
Multi-domain spoken dialogue system with extensibility and robustness against speech recognition errors

SigDIAL '06 Proceedings of the 7th SIGdial Workshop on Discourse and Dialogue
A multi-expert model for dialogue and behavior control of conversational robots and agents

Knowledge-Based Systems
A classifier-based approach to supporting the augmentation of the question-answer database for spoken dialogue systems

IWSDS'10 Proceedings of the Second international conference on Spoken dialogue systems for ambient environments

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes a general and effective domain selection framework for multi-domain spoken dialogue systems that employ distributed domain experts. The framework consists of two processes: deciding if the current domain continues and estimating the probabilities for selecting other domains. If the current domain does not continue, the domain with the highest activation probability is selected. Since those processes for each domain expert can be designed independently from other experts and can use a large variety of information, the framework achieves both extensibility and robustness against speech recognition errors. The results of an experiment using a corpus of dialogues between humans and a multi-domain dialogue system demonstrate the viability of the proposed framework.