ABSTRACT
Several studies have investigated the effect of emotions on the overall behavior and learning skills of children. Emotions are extremely important in determining mental health and environmental pressure of an individual. Emotions are an expression of one's perspective or one's mental state to others. Speech emotion recognition (SER) is the process of recognizing individual emotions and affective states from speech data. It can predict basic emotions just by analyzing the speech of the speaker, which could help the speech and reading skills of learning-disabled learners. Therefore, in the present study the MLPClassifier model is used to predict the emotion of dyslexic learners to identify their problematic skills. Here, varied pitched voices are used to achieve the objective. The potential features are extracted from each utterance for the computational mapping between emotions and speech patterns. The pitch is analyzed from the selected features to discriminate dyslexic and non-dyslexic learners. A total of 19 learning- and non-learning-disabled learners have been included in this study to discriminate their recorded audios through SER. Also, we have observed how learning-disabled learners' emotions get influenced in personal and academic areas of their life.
