Affective computing
Infant-like social interactions between a robot and a human caregiver
Adaptive Behavior
The production and recognition of emotions in speech: features and algorithms
International Journal of Human-Computer Studies - Application of affective computing in humanComputer interaction
Detecting changing emotions in natural speech
IEA/AIE'12 Proceedings of the 25th international conference on Industrial Engineering and Other Applications of Applied Intelligent Systems: advanced research in applied artificial intelligence
Detecting changing emotions in human speech by machine and humans
Applied Intelligence
Hi-index | 0.00 |
Estimation of emotion in speech is one of issue to enhance human computer interaction. Some researchers tried to estimate emotion in speech and there is a conventional approach. However, the performance of it is not good. In order to enhance the conventional approach for estimation emotion in speech, we added a new step to the conventional approach. The new step is that we subdivide speech data before extracting relations for estimation emotion in speech. We use a nonhierarchical cluster analysis to subdivide speech data. We designed this approach based on the idea that there are various expressions for one emotion. In order to evaluate the useful of our approach, we did experiments. Experimental results show that our approach is better than the conventional approach. However, the accuracy rate and the number of correct answers are still low. We should continue to improve our approach. We have some future works. One of them is to raise accuracy of subdividing. Therefore, we are trying to define criteria and make a new cluster analysis.