Bimodal Emotion Recognition from Speech and Text
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2014, Vol 5, Issue 2
Abstract
This paper presents an approach to emotion recognition from speech signals and textual content. In the analysis of speech signals, thirty-seven acoustic features are extracted from the speech input. Two different classifiers Support Vector Machines (SVMs) and BP neural network are adopted to classify the emotional states. In text analysis, we use the two-step classification method to recognize the emotional states. The final emotional state is determined based on the emotion outputs from the acoustic and textual analyses. In this paper we have two parallel classifiers for acoustic information and two serial classifiers for textual information, and a final decision is made by combing these classifiers in decision level fusion. Experimental results show that the emotion recognition accuracy of the integrated system is better than that of either of the two individual approaches.
Authors and Affiliations
Weilin Ye, Xinghua Fan
Keywords
Related Articles
- EP ID EP104871
- DOI 10.14569/IJACSA.2014.050204
- Views 127
- Downloads 0
A Personalized Hybrid Recommendation Procedure for Internet Shopping Support
Lately, recommender systems (RS) have offered a remarkable breakthrough to users. It lessens the user time cost thereby delivering faster and better results. After purchasing a product there are recommendations according...
A Block Cipher Involving a Key and a Key Bunch Matrix, Supplemented with Key-Based Permutation and Substitution
In this paper, we have developed a block cipher involving a key and a key bunch matrix. In this cipher, we have made use of key-based permutation and key-based substitution. The cryptanalysis carried out in this investig...
Analysis of Software Reliability Data using Exponential Power Model
In this paper, Exponential Power (EP) model is proposed to analyze the software reliability data and the present work is an attempt to represent that the model is as software reliability model. The approximate MLE using...
A Novel Approach for Dimensionality Reduction and Classification of Hyperspectral Images based on Normalized Synergy
During the last decade, hyperspectral images have attracted increasing interest from researchers worldwide. They provide more detailed information about an observed area and allow an accurate target detection and precise...
Developing an Algorithm for Securing the Biometric Data Template in the Database
In the current technology advancement, biometric template provides a dependable solution to the problem of user verification in an identity control system. The template is saved in the database during the enrollment and...