sorry, we can't preview this file

...but you can still download file.pdf
file.pdf (61.58 kB)

You Are What You Say: Using Meeting Participants’ Speech to Detect their Roles and Expertise

Download (61.58 kB)
journal contribution
posted on 01.01.2005, 00:00 by Satanjeev Banerjee, Alexander RudnickyAlexander Rudnicky

Our goal is to automatically detect the functional roles that meeting participants play, as well as the expertise they bring to meetings. To perform this task, we build decision tree classifiers that use a combination of simple speech features (speech lengths and spoken keywords) extracted from the participants’ speech in meetings. We show that this algorithm results in a role detection accuracy of 83% on unseen test data, where the random baseline is 33.3%. We also introduce a simple aggregation mechanism that combines evidence of the participants’ expertise from multiple meetings. We show that this aggregation mechanism improves the role detection accuracy from 66.7% (when aggregating over a single meeting) to 83% (when aggregating over 5 meetings).

History

Publisher Statement

All Rights Reserved

Date

01/01/2005

Usage metrics

Exports