Sixth International Conference on Spoken Language Processing
This paper analysis the feature of the time amplitude pitch and formant construction involved such four emotions as happiness anger surprise and sorrow. Through comparison with non-emotional speech signal, we sum up the distribution law of emotional feature including different emotional speech. Nine emotional features were extracted from emotional speech for recognizing emotion. We introduce three emotional recognition methods based on principal component analysis and the results show that these method can provide an effective solution to emotional recognition.
Bibliographic reference. Zhao, Li / Lu, Wei / Jiang, Ye / Wu, Zhenyang (2000): "A study on emotional feature recognition in speech", In ICSLP-2000, vol.3, 961-964.