Speech emotion recognition on smart computer : can you tell the mood ?

• Main Author: Shao, Qing
• Other Authors: Khong Andy Wai Hoong
• Format: Final Year Project
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/63836
• Institution: Nanyang Technological University

Speech emotion recognition is a technology, which could recognize the emotion state of speakers based on the speech signals. The development of a speech emotion recognition system is demonstrated in this project. The system is designed on MATLAB program based on Artificial Neural Network (ANN) algorithm. The model is developed and improved step by step by model selection. In the model selection process, the impact of the different parameters and speech features will be studied. After a set of tests based on the changing of parameters and various combinations of speech features, the model with best performance is selected. Finding out best value of parameters and the most significant features, which have the greatest impact on speech emotion recognition, could improve the accuracy and also reduce the processing time.