Automatic Detection of Nominal Entities in Speech for Enriched Content Search

ORCiD

Leili Javadpour: 0000-0003-4004-1950

Document Type

Conference Presentation

Conference Title

Twenty-Sixth International Florida Artificial Intelligence Research Society Conference (FLAIRS-26)

Location

St. Pete Beach, FL

Conference Dates

May 22-24, 2013

Date of Presentation

5-24-2013

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.

This document is currently not available here.

Share

COinS