Abstract- The analysis of human speech is a very challengingresearch area as it concerns the detection of user communities. Emotions playan initial role in human interaction. The ability to understand users emotionsby analyzing voice is desirable in different applications of speech recognitionin emotions can be found in different areas, such as the interaction betweencomputers and humans and call centers. Previously, emotion recognition made useof simple classifiers on bag-of-words models. However, the existing work ofemotion recognition on Voice was carried out with the help of deep learningtechniques on static voice data. The proposed method focuses on increasing theoverall accuracy of emotion detection during calls using artificialintelligence. The overall aim is to accurately recognize the various emotionsthat a particular speech expresses semantically.
Keywords-Emotion, feature extraction, Artificial intelligence, Speech to Text.