acl acl2013 acl2013-124 acl2013-124-reference knowledge-graph by maker-knowledge-mining

124 acl-2013-Discriminative state tracking for spoken dialog systems


Source: pdf

Author: Angeliki Metallinou ; Dan Bohus ; Jason Williams

Abstract: In spoken dialog systems, statistical state tracking aims to improve robustness to speech recognition errors by tracking a posterior distribution over hidden dialog states. Current approaches based on generative or discriminative models have different but important shortcomings that limit their accuracy. In this paper we discuss these limitations and introduce a new approach for discriminative state tracking that overcomes them by leveraging the problem structure. An offline evaluation with dialog data collected from real users shows improvements in both state tracking accuracy and the quality of the posterior probabilities. Features that encode speech recognition error patterns are particularly helpful, and training requires rel- atively few dialogs.


reference text

AT&T; Statistical Dialog Toolkit. AT&T; Statistical Dialog Toolkit. http : / /www2 . re s earch . att . com/ sw/ t oo l / a s dt / , 2013. s Adam Berger, Vincent J. Della Pietra, and Stephen A. Della Pietra. A maximum entropy approach to natural language processing. Computational Linguistics, 22:39–71, 1996. 474 Alan W. Black, S. Burger, B. Langner, G. Par- ent, and M. Eskenazi. Spoken dialog challenge 2010. In Proc. of Workshop on Spoken Language Technologies (SLT), 2010. Dan Bohus and Alex Rudnicky. A k hypotheses + other belief updating model. In Proc. of AAAI Workshop on Statistical and Empirical Approaches to Spoken Dialog Systems, 2006. Milica Gaˇ si´ c and Steve Young. Effective handling of dialogue state in the hidden information state pomdp dialogue manager. ACM Transactions on Speech and Language Processing, 7, 2011. Eric Horvitz and Tim Paek. A computational architecture for conversation. In Proc. of the 7th Intl. Conf. on User Modeling, 1999. Allan H Murphy. A new vector partition of the probability score. Journal of Applied Meteorology, 12:595–600, 1973. Blaise Thomson and Steve Young. Bayesian update of dialogue state: A POMDP framework for spoken dialogue systems. Computer Speech and Language, 24(4):562–588, 2010. Jason D. Williams. Incremental partition recombination for efficient tracking of multiple dialogue states. In Proc. of ICASSP, 2010. Jason D. Williams. Challenges and opportunities for state tracking in statistical spoken dialog systems: Results from two public deployments. IEEE Journal of Selected Topics in Signal Processing, Special Issue on Advances in Spoken Dialogue Systems and Mobile Interface, 6(8): 959–970, 2012. Jason D. Williams and Suhrid Balakrishnan. Estimating probability of correctness for asr n-best lists. In Proc. SigDial Conference, 2009. Jason D. Williams and Steve Young. Partially observable markov decision processes for spoken dialog systems. Computer Speech and Language, 21:393–422, 2007. Jason D. Williams, Iker Arizmendi, and Alistair Conkie. Demonstration of AT&T; Let’s Go: A production-grade statistical spoken dialog system. In Proc of Workshop on Spoken Language Technologies (SLT), 2010. Steve Young, Milica Gaˇ si´ c, Simon Keizer, Fran ¸cois Mairesse, Jost Schatzmann, Blaise Thomson, and Kai Yu. The hidden informa- tion state model: a practical framework for POMDP-based spoken dialogue management. Computer Speech and Language, 24(2): 150– 174, 2010. Bianca Zadrozny and Charles Elkan. Transforming classifier scores into accurate multiclass probability estimates. In Proc. of the eighth ACM SIGKDD Intl. Conf on Knowledge Discovery and Data mining, pages 694–699, 2002. 475