Survey on human speech emotions identification using deep learning algorithms

Emotion recognition plays an important role in human-computer intercommunication, in the medical field, in call centers, etc. Gradually, the chances of Emotion Computing are increasing. The main six primary emotions are anger, disgust, fear, joy, sadness, and surprise. Many types of emotion recognition methods are available to recognize these emotions. But they mainly focus on facial expressions, speech, and gestures. Using visible emotion cues cannot capture people’s true emotions. To get real emotions, emotion recognition using a Convolutional neural network (CNN), Recurrent neural network (RNN), and support vector machine (SVM), LSTM becomes essential. This article discusses various methods used for emotion recognition using RNN, LSTM using machine learning and deep learning techniques, data mining methods, and their databases. It also includes the experimental results and accuracy of each method. With this focus on building an emotionless machine, brilliance can be achieved.