Edinburgh Research Explorer

Dynamic Bayesian networks for meeting structuring

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Original languageEnglish
Title of host publicationAcoustics, Speech, and Signal Processing, 2004
Subtitle of host publicationProceedings. (ICASSP '04). IEEE International Conference on (Volume:5 )
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages629-632
DOIs
Publication statusPublished - 2004
Event2004 IEEE International Conference on Acoustics, Speech and Signal Processing (IEEE ICASSP) - Fairmount Queen Elizabeth Hotel, Montreal, Quebec, Canada
Duration: 17 May 200421 May 2004

Conference

Conference2004 IEEE International Conference on Acoustics, Speech and Signal Processing (IEEE ICASSP)
CountryCanada
CityMontreal, Quebec
Period17/05/0421/05/04

Abstract

The paper is about the automatic structuring of multiparty meetings using audio information. We have used a corpus of 53 meetings, recorded using a microphone array and lapel microphones for each participant. The task was to segment meetings into a sequence of meeting actions, or phases. We have adopted a statistical approach using dynamic Bayesian networks (DBNs). Two DBN architectures were investigated: a two-level hidden Markov model (HMM) in which the acoustic observations were concatenated; and a multistream DBN in which two separate observation sequences were modelled. We have also explored the use of counter variables to constrain the number of action transitions. Experimental results indicate that the DBN architectures are an improvement over a simple baseline HMM, with the multistream DBN with counter constraints producing an action error rate of 6%.

Event

2004 IEEE International Conference on Acoustics, Speech and Signal Processing (IEEE ICASSP)

17/05/0421/05/04

Montreal, Quebec, Canada

Event: Conference

Download statistics

No data available

ID: 27411401