Emotion recognition in Filipino speech: EMOTICON

Date of Publication

2009

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Computer Science

College

College of Computer Studies

Department/Unit

Computer Science

Defense Panel Member

Charibeth Ko Cheng
Merlin Teodosia C. Suarez
Rafael A. Cabredo

Abstract/Summary

Accurate recognition of emotions in a given speech has a great benefit in the speech interfaces between human and computers. It adds to the appeal of electronic systems by contributing to the user's perception of the system's intelligence and adaptability. However, feature extraction and algorithms are still disputed issues for the recognition of emotions and existing systems are having issues in terms of accuracy when applied with other languages such as the Filipino language. This paper proposes a system capable of recognizing different emotional states based on the Filipino language utterances. The system identifies acoustic features that correlate to attain the following emotional states: happiness, sadness, anger, fear, surprise, disgust and neutral. Algorithms of existing emotion recognition systems were used as guide to determine the appropriate algorithms and features that should be used to yield higher accuracy. The emotional classifier was implemented using linear search to locate the K-nearest neighbors. This classifier worked by getting the Euclidean distances between two feature vectors and classifying the input's emotion based on its nearest neighbors. The system extracted a minimal acoustic feature set that uniquely identified each emotion. Pitch, energy, duration, and formants were the acoustic features extracted. Among these, pitch and energy were used as the minimal acoustic feature set based on the tests conducted. Using good quality speech samples and the minimal feature set, the system was able to produce a recognition accuracy of 40.12%.

Abstract Format

html

Language

English

Format

Print

Accession Number

TU15433

Shelf Location

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

Physical Description

1 v. (various foliations) : ill. (some col.) 29 cm

Keywords

Speech processing systems--Computer programs; Automatic speech recognition--Data processing

This document is currently not available here.

Share

COinS