Title

Investigating the use of music features in audio-based emotion detection in laughter

Date of Publication

2013

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Computer Science

Subject Categories

Computer Sciences

College

College of Computer Studies

Department/Unit

Computer Science

Thesis Adviser

Jocelynn Cu

Defense Panel Member

Merlin Teodisia C. Suarez

Macaria O. Cordel, II

Abstract/Summary

Laughter is one of the pan-human expressive acts. It is a powerful affective and social signal since people very often express their emotion and regulate conversations by laughing. Current works on laughter focus on analyzing it through the use of spectral and prosodic features. The differentiating factor in this work is the use of music features. We investigate the usefulness of using music features for the analysis of laughter and discrimination of emotion. We perform different building, feature extraction and modeling and validation. All of this was geared towards determining the effectiveness of music features. Results indicate that while prosodic and spectral features are good in discriminating emotions, music features does an equal and/or even better job, showing its usefulness in this area of computing.

Abstract Format

html

Language

English

Format

Print

Accession Number

TU18491

Shelf Location

Archives, The Learning Commons, 12F, Henry Sy Sr. Hall

Physical Description

1v. various foliations : illustrations (some colored) ; 28 cm.

Keywords

Laughter; Philippine wit and humor

This document is currently not available here.

Share

COinS