Publication View

Gestural Cohesion for Topic Segmentation (2009)

Abstract
This paper explores the relationship between discourse segmentation and coverbal gesture. Introducing the idea of gestural cohesion, we show that coherent topic segments are characterized by homogeneous gestural forms and that changes in the distribution of gestural features predict segment boundaries. Gestural features are extracted automatically from video, and are combined with lexical features in a Bayesian generative model. The resulting multimodal system outperforms text-only segmentation on both manual and automaticallyrecognized speech transcripts. 1

Publication details
Download http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.146.796
Source http://people.csail.mit.edu/jacobe/papers/GCTS_ACL2008.pdf
Contributors CiteSeerX
Repository CiteSeerX - Scientific Literature Digital Library and Search Engine (United States)
Type text
Language English
Relation 10.1.1.131.2084, 10.1.1.14.9528, 10.1.1.21.1181, 10.1.1.131.2072, 10.1.1.15.205, 10.1.1.60.2391, 10.1.1.37.8065, 10.1.1.112.3591, 10.1.1.14.5261, 10.1.1.3.8962, 10.1.1.77.5712, 10.1.1.135.6194, 10.1.1.144.4356, 10.1.1.19.1004, 10.1.1.14.7542, 10.1.1.143.4245, 10.1.1.126.5521, 10.1.1.130.7963