Automatic detection of nominal entities in speech for enriched content search

Document Type

Conference Proceeding

Publication Date

12-13-2013

Abstract

In this work, a methodology is developed to detect sentient actors in spoken stories. Meta-tags are then saved to XML files associated with the audio files. A recursive approach is used to find actor candidates and features which are then classified using machine learning approaches. Results of the study indicate that the methodology performed well on a narrative based corpus of children's stories. Using Support Vector Machines for classification, an F-measure accuracy score of 86% was achieved for both named and unnamed entities. Additionally, feature analysis indicated that speech features were very useful when detecting unnamed actors.

Publication Source (Journal or Book title)

FLAIRS 2013 - Proceedings of the 26th International Florida Artificial Intelligence Research Society Conference

First Page

190

Last Page

195

This document is currently not available here.

Share

COinS