A repository for researching and developing SotA models for Speech Emotion Recognition (SER)
Speech emotion recognition (SER) is the act of attempting to recognize human emotion and affective states from speech. In recent years, deep architectures have achieved state-of-the-art performance for many of the most prominent SER datasets. In this repository, I explore some of the most cutting-edge models as I venture to research and develop possible improvements.
The Berlin Database for Emotional Speech (Emo-DB) consists of over 500 short audio clips capturing emotional utterances by voice actors. Each audio clip depicts one of seven emotional classes (Anger, Boredom, Disgust, Fear, Happiness, Sadness, and Neutral). The Emo-DB dataset is a commonly-used benchmark for evaluating computational SER approaches.