Research on Speech Emotion Recognition in E-learning By Using Neural Networks Method

Emotion deficiency was a hot topic in present e-learning research. The author analyzed many negative effects of emotion deficiency and proposed a lot of corresponding countermeasures. Basing on it, affective computing was applied in the traditional e-learning system. The model of e-learning system based on affective computing was constructed by using speech emotion, which took speech feature as input data. Neural networks method was used to speech emotion recognition. Our simulation experiment results showed that neural networks method was effective in emotion recognition, and we achieved a recognition rate of approximately 55% when testing eight emotions.