论文信息 - Asking Questions to Limited Domain Virtual Characters: How Good Does Speech Recognition Have to Be?

Asking Questions to Limited Domain Virtual Characters: How Good Does Speech Recognition Have to Be?

In this paper, we describe the evaluation of a limited domain question-answering characters, particularly as to the effect of non-optimal speech recognition, and the ability to appropriately answer novel questions. Results show that answering ability is robust until speech recognition reaches over 60% Word error rate.

[1] Marilyn A. Walker,et al. Design and evaluation of spoken dialog systems , 1997, 1997 IEEE Workshop on Automatic Speech Recognition and Understanding Proceedings.

[2] Anton Leuski. Email is a stage: discovering people roles from email archives , 2004, SIGIR '04.

[3] Diane J. Litman,et al. Correlating student acoustic-prosodic profiles with student learning in spoken tutoring dialogues , 2005, INTERSPEECH.

[4] W. Bruce Croft,et al. Text Segmentation by Topic , 1997, ECDL.

[5] David R. Traum,et al. Improving question-answering with linking dialogues , 2006, IUI '06.

[6] Anton Leuski,et al. Dealing with Out of Domain Questions in Virtual Characters , 2006, IVA.

[7] Anton Leuski,et al. How to talk to a hologram , 2006, IUI '06.

[8] James P. Callan,et al. Training algorithms for linear text classifiers , 1996, SIGIR '96.

[9] Panayiotis G. Georgiou,et al. Building topic specific language models from webdata using competitive models , 2005, INTERSPEECH.

[10] Bob Carpenter,et al. Vector-based Natural Language Call Routing , 1999, Comput. Linguistics.

[11] Joseph Weizenbaum,et al. and Machine , 1977 .

[12] Anton Leuski,et al. Building Effective Question Answering Characters , 2006, SIGDIAL Workshop.

[13] Thomas Hofmann,et al. Support vector machine learning for interdependent and structured output spaces , 2004, ICML.

[14] Jimmy J. Lin,et al. Overview of the TREC 2007 Question Answering Track , 2008, TREC.