Abstract
To solve the speaker independent emotion recognition problem, a multi-level speech emotion recognition system is proposed to classify 6 speech emotions, including sadness, anger, surprise, fear, happiness and disgust from coarse to fine. The key is that the emotions divided by each layer are closely related to the emotional features of speech. For each level, appropriate features are selected from 288 candidate features by Fisher ratio which is also regarded as input parameter for the training of support vector machine (SVM). Based on Beihang emotional speech database and Berlin emotional speech database, principal component analysis (PCA) for dimension reduction and Artificial Neural Network (ANN) for classification are adopted to design 4 comparative experiments, including Fisher+SVM, PCA+SVM, Fisher+ANN, PCA+ANN. The experimental results prove that Fisher rule is better than PCA for dimension reduction, and SVM is more expansible than ANN for speaker independent speech emotion recognition. Good cross-cultural adaptation can be inferred from the similar results of experiments based on two different databases.
| Original language | English |
|---|---|
| Pages (from-to) | 604-609 |
| Number of pages | 6 |
| Journal | Moshi Shibie yu Rengong Zhineng/Pattern Recognition and Artificial Intelligence |
| Volume | 25 |
| Issue number | 4 |
| Publication status | Published - 2012 Aug |
| Externally published | Yes |
Fingerprint
Keywords
- Fisher criterion
- Speaker independent
- Speech emotion recognition
- Support vector machine
ASJC Scopus subject areas
- Artificial Intelligence
- Computer Vision and Pattern Recognition
- Software
Cite this
Multi-level speech emotion recognition based on Fisher criterion and SVM. / Chen, Li Jiang; Mao, Xia; Ishizuka, Mitsuru.
In: Moshi Shibie yu Rengong Zhineng/Pattern Recognition and Artificial Intelligence, Vol. 25, No. 4, 08.2012, p. 604-609.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Multi-level speech emotion recognition based on Fisher criterion and SVM
AU - Chen, Li Jiang
AU - Mao, Xia
AU - Ishizuka, Mitsuru
PY - 2012/8
Y1 - 2012/8
N2 - To solve the speaker independent emotion recognition problem, a multi-level speech emotion recognition system is proposed to classify 6 speech emotions, including sadness, anger, surprise, fear, happiness and disgust from coarse to fine. The key is that the emotions divided by each layer are closely related to the emotional features of speech. For each level, appropriate features are selected from 288 candidate features by Fisher ratio which is also regarded as input parameter for the training of support vector machine (SVM). Based on Beihang emotional speech database and Berlin emotional speech database, principal component analysis (PCA) for dimension reduction and Artificial Neural Network (ANN) for classification are adopted to design 4 comparative experiments, including Fisher+SVM, PCA+SVM, Fisher+ANN, PCA+ANN. The experimental results prove that Fisher rule is better than PCA for dimension reduction, and SVM is more expansible than ANN for speaker independent speech emotion recognition. Good cross-cultural adaptation can be inferred from the similar results of experiments based on two different databases.
AB - To solve the speaker independent emotion recognition problem, a multi-level speech emotion recognition system is proposed to classify 6 speech emotions, including sadness, anger, surprise, fear, happiness and disgust from coarse to fine. The key is that the emotions divided by each layer are closely related to the emotional features of speech. For each level, appropriate features are selected from 288 candidate features by Fisher ratio which is also regarded as input parameter for the training of support vector machine (SVM). Based on Beihang emotional speech database and Berlin emotional speech database, principal component analysis (PCA) for dimension reduction and Artificial Neural Network (ANN) for classification are adopted to design 4 comparative experiments, including Fisher+SVM, PCA+SVM, Fisher+ANN, PCA+ANN. The experimental results prove that Fisher rule is better than PCA for dimension reduction, and SVM is more expansible than ANN for speaker independent speech emotion recognition. Good cross-cultural adaptation can be inferred from the similar results of experiments based on two different databases.
KW - Fisher criterion
KW - Speaker independent
KW - Speech emotion recognition
KW - Support vector machine
UR - http://www.scopus.com/inward/record.url?scp=84867418522&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84867418522&partnerID=8YFLogxK
M3 - Article
VL - 25
SP - 604
EP - 609
JO - Moshi Shibie yu Rengong Zhineng/Pattern Recognition and Artificial Intelligence
JF - Moshi Shibie yu Rengong Zhineng/Pattern Recognition and Artificial Intelligence
SN - 1003-6059
IS - 4
ER -