Multistream Recognition of Dialogue Acts in Meetings

Alfred Dielmann, Steve Renals

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

We propose a joint segmentation and classification approach for the dialogue act recognition task on natural multi-party meetings (ICSI Meeting Corpus). Five broad DA categories are automatically recognised using a generative Dynamic Bayesian Network based infrastructure. Prosodic features and a switching graphical model are used to estimate DA boundaries, in conjunction with a factored language model which is used to relate words and DA categories. This easily generalizable and extensible system promotes a rational approach to the joint DA segmentation and recognition task, and is capable of good recognition performance.
Original languageEnglish
Title of host publicationMachine Learning for Multimodal Interaction
Subtitle of host publicationThird International Workshop, MLMI 2006, Bethesda, MD, USA, May 1-4, 2006, Revised Selected Papers
EditorsSteve Renals, Samy Bengio, Jonathan G. Fiscus
Place of PublicationBerlin, Heidelberg
PublisherSpringer Berlin Heidelberg
Pages178-189
Number of pages12
ISBN (Electronic)978-3-540-69268-3
ISBN (Print)978-3-540-69267-6
DOIs
Publication statusPublished - 2006
EventThird International Workshop MLMI 2006 - Bethesda, MD, United States
Duration: 1 May 20064 May 2006

Publication series

NameLecture Notes in Computer Science
PublisherSpringer Berlin Heidelberg
Volume4299
ISSN (Print)0302-9743

Workshop

WorkshopThird International Workshop MLMI 2006
CountryUnited States
CityBethesda, MD
Period1/05/064/05/06

Fingerprint Dive into the research topics of 'Multistream Recognition of Dialogue Acts in Meetings'. Together they form a unique fingerprint.

Cite this