Error simulation for training statistical dialogue systems

被引：27

作者：

Schatzmann, Jost ^{[1
]}

Thomson, Blaise ^{[1
]}

Young, Steve ^{[1
]}

机构：

[1] Univ Cambridge, Dept Engn, Cambridge CB2 1PZ, England

来源：

2007 IEEE WORKSHOP ON AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING, VOLS 1 AND 2 | 2007年

关键词：

error simulation; statistical modelling; spoken dialogue systems; PO MP; dialogue policy training;

D O I：

10.1109/ASRU.2007.4430167

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Human-machine dialogue is heavily influenced by speech recogni tion and understanding errors and it is hence desirable to train and test statistical dialogue system policies under realistic noise conditions. This paper presents a novel approach to error simulation based on statistical models for word-level utterance generation, ASR confusions, and confidence score generation. While the method explicitly models the context-dependent acoustic confusability of words and allows the system specific language model and semantic decoder to be incorporated, it is computationally inexpensive and thus potentially suitable for running thousands of training simulations. Experimental evaluation results with a POMDP-based dialogue system and the Hidden Agenda User Simulator indicate a close match between the statistical properties of real and synthetic errors.

引用

页码：526 / 531

页数：6