Skip to main navigation Skip to search Skip to main content

Training a sentence planner for spoken dialog: The impact of syntactic and planning features

  • Carnegie Mellon University
  • AT&T

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

The dialog manager of a spoken dialog system often performs domain dependent functions as well as general dialog tasks. It is possible to separate the domain specific knowledge from knowledge about language using techniques from natural language generation. However a natural language generator often has to be tuned for particular applications. In this work, we describe a new method for automatically training the natural language generator and examine the role that domain specific and domain independent features have on performance. We show that although the general features have the largest impact, the use of domain specific features improves performance, while still retaining the benefits of automatic domain customization through training.

Original languageEnglish
Title of host publicationEUROSPEECH 2001 - SCANDINAVIA - 7th European Conference on Speech Communication and Technology
EditorsBorge Lindberg, Henrik Benner, Paul Dalsgaard, Zheng-Hua Tan
PublisherInternational Speech Communication Association
Pages1747-1750
Number of pages4
ISBN (Electronic)8790834100, 9788790834104
StatePublished - 2001
Event7th European Conference on Speech Communication and Technology - Scandinavia, EUROSPEECH 2001 - Aalborg, Denmark
Duration: Sep 3 2001Sep 7 2001

Publication series

NameEUROSPEECH 2001 - SCANDINAVIA - 7th European Conference on Speech Communication and Technology

Conference

Conference7th European Conference on Speech Communication and Technology - Scandinavia, EUROSPEECH 2001
Country/TerritoryDenmark
CityAalborg
Period09/3/0109/7/01

Fingerprint

Dive into the research topics of 'Training a sentence planner for spoken dialog: The impact of syntactic and planning features'. Together they form a unique fingerprint.

Cite this