Investigation on dynamic speech emotion from the perspective of brain associative memory.
Many researchers have studied speech emotion for years from the perspective of psychology to engineering. To date, none has made the speech emotion recognition system intuitive enough in such a way that it can be embedded in automatic answering machines that can effectively detect the various affect...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2013
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/38078/ http://irep.iium.edu.my/38078/ http://irep.iium.edu.my/38078/1/Investigation_on_Dynamic_Speech_Emotion_from_the_Perspective_of_Brain_Associative_Memory.pdf |
| Summary: | Many researchers have studied speech emotion for years from the perspective of psychology to engineering. To date, none has made the speech emotion recognition system intuitive enough in such a way that it can be embedded in automatic answering machines that can effectively detect the various affective states of human verbal communication. In most cases the underlying emotional information was misinterpreted thus resulting in wrong feedbacks and responses. The complexity of understanding and analyzing speech emotion is presented in the dynamics of the emotion itself. Emotion is dynamic and changeable over time. Hence, it is imperative to cater for this parameter to boost the performance of the speech emotion recognition system. In this paper, values of Valence (V) and Arousal(A) are used to generate a recalibrated affective space model. Such approach is adopted from psychologists' understanding that emotion can be represented using emotion primitives' values. The VA approach is then coupled with the brain associative memory concept that can provides a better means in understanding the dynamics of speech emotion. Results of such analysis tallies with the psychological findings and has its practical implementation. |
|---|