Speech emotion recognition: Features and classification models

To solve the speaker independent emotion recognition problem, a three-level speech emotion recognition model is proposed to classify six speech emotions, including sadness, anger, surprise, fear, happiness and disgust from coarse to fine. For each level, appropriate features are selected from 288 ca...

Full description

Saved in:
Bibliographic Details
Published in:Digital signal processing 2012-12, Vol.22 (6), p.1154-1160
Main Authors: Chen, Lijiang, Mao, Xia, Xue, Yuli, Cheng, Lee Lung
Format: Article
Language:English
Subjects:
Classification
Emotion recognition
Emotions
Fisher discriminant
Learning theory
Neural networks
Recognition
Speaker independent
Speech
Speech recognition
Support vector machines
SVM
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:To solve the speaker independent emotion recognition problem, a three-level speech emotion recognition model is proposed to classify six speech emotions, including sadness, anger, surprise, fear, happiness and disgust from coarse to fine. For each level, appropriate features are selected from 288 candidates by using Fisher rate which is also regarded as input parameter for Support Vector Machine (SVM). In order to evaluate the proposed system, principal component analysis (PCA) for dimension reduction and artificial neural network (ANN) for classification are adopted to design four comparative experiments, including Fisher+SVM, PCA+SVM, Fisher+ANN, PCA+ANN. The experimental results proved that Fisher is better than PCA for dimension reduction, and SVM is more expansible than ANN for speaker independent speech emotion recognition. The average recognition rates for each level are 86.5%, 68.5% and 50.2% respectively.
ISSN:1051-2004
1095-4333
DOI:10.1016/j.dsp.2012.05.007