A data-driven approach to detection of interruptions in human-human conversations

Raveesh Meena, Saeed Dabbaghchian, Kalin Stefanov

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

Abstract

We report the results of our initial ef- forts towards automatic detection of user’s interruptions in a spoken human– machine dialogue. In a first step, we explored the use of automatically ex- tractable acoustic features, frequency and intensity, in discriminating listen- er’s interruptions in human–human conversations. A preliminary analysis of interaction snippets from the HCRC Map Task corpus suggests that for the task at hand, intensity is a stronger fea- ture than frequency, and using intensity in combination with feature loudness offers the best results for a k-means clustering algorithm
Original languageEnglish
Title of host publicationProceedings from FONETIK 2014 Stockholm, June 9-11, 2014
EditorsMattias Heldner
Place of PublicationStockholm Sweden
PublisherStockholm University
Pages29-32
Number of pages4
ISBN (Electronic)9789163756634
ISBN (Print)9789163756627
Publication statusPublished - 2014

Publication series

NameProceedings from FONETIK 2014 Stockholm, June 9-11, 2014 PERILUS XXIV, June 2014
PublisherStockholm University
ISSN (Electronic)0282-6690

Cite this