A Study on Content-Based Classification and Retrieval of Audio Database
IDEAS '01 Proceedings of the International Database Engineering & Applications Symposium
Probability Estimates for Multi-class Classification by Pairwise Coupling
The Journal of Machine Learning Research
Emotion Recognition and Synthesis System on Speech
ICMCS '99 Proceedings of the IEEE International Conference on Multimedia Computing and Systems - Volume 2
Knowledge Engineering for Affective Bi-Modal Interaction in Mobile Devices
Proceedings of the 2008 conference on Knowledge-Based Software Engineering: Proceedings of the Eighth Joint Conference on Knowledge-Based Software Engineering
Affective speaker state analysis in the presence of reverberation
International Journal of Speech Technology
Multimodal object oriented user interfaces in mobile affective interaction
Multimedia Tools and Applications
Robust emotional speech classification in the presence of babble noise
International Journal of Speech Technology
Hi-index | 0.00 |
In this paper, a speech emotion recognition agent for mobile communication service is proposed. The proposed system can recognize five emotional states - neutral, happiness, sadness, anger, and annoyance from the speech captured by a cellular phone in real time and then it calculates the degree of affection such as love, truthfulness, weariness, trick, friendship of the person who you are interesting to know through the mobile phone. In general, a speech acquired by a cellular phone contains noise due to the mobile network and environmental noise. Thus it can causes serious performance degradation due to the distortion in emotional features of the query speech. In order to alleviate the effect of these noises, we adopt a MA (Moving Average) filter which has relatively simple structure and low computational complexity. Then a feature optimization method is implemented to further improve and stabilize the system performance. For a practical application, we create an agent that can measure the degree of affection from the person who you want to know on the mobile phone. Two pattern classification methods, k-NN and SVM with probability estimates, are compared for estimating the degree of affection. The experimental results indicate that the proposed method provides very stable and successful emotional classification performance as 72.5 % over five emotional states and it shows the feasibility of the agent for mobile communication services.