Emotion Estimation by Joint Facial Expression and Speech Tonality Using Evolutionary Deep Learning Structures








Abstract

This work proposes an emotion recognition system by adopting facial expression and speech tonality on deep learning networks. Both convolutional neural networks and long short term memory networks are used for feature training. The two features can be trained together to acquire higher accuracy. Moreover, Structure Evolution which is inspired by the Genetic Algorithm is added to optimize the parameters in the model. The experimental results show the joint model optimized by Structure Evolution surpasses the single model by at least 10% and outperforms the state-of-the-art work over 1%.


Modules


Algorithms

Machine learning algorithms


Software And Hardware

• Hardware: Processor: i3 ,i5 RAM: 4GB Hard disk: 16 GB • Software: operating System : Windws2000/XP/7/8/10 Anaconda,jupyter,spyder,flask Frontend :-python Backend:- MYSQL