Speech Emotion Recognition from Audio Files Using Feedforward Neural Network

Speech Emotion Recognition (SER) is an act of recognizing human emotion and affective states from speech. This is because voice often reflects underlying emotion through tone and pitch. Emotion recognition is a component of speech recognition which is gaining more popularity and need for it is increasing exceptionally. Our paper aims to use the Feedforward Neural Network to recognize emotions from unseen data (i.e., audio files) and label them according to the range of different emotions using appropriate variables (such as modality, emotion, intensity, repetition etc.) found in the data. This approach can be particularly useful in the case of identifying the inherent emotion behind human voice.