Emotion Recognition based on Human voice tones

Abstract

In this period of technology and automation, human interaction with machines has become an unavoidable occurrence. Machines are making our lives much easier, stretching their hands wherever it is possible for smoother operation of various tasks. Emotion recognition system is one of the system that can help humans for a much easier and meaningful interaction with machines. It can be used in various branches such as artificial intelligence, health care, psychology, law, e-learning, marketing, cognitive science, entertainment and many more. The human voice tones are very versatile and carries a multitude of emotions. Emotion in speech carries extra insight about human actions. Through deeper analysis, we can better understand the motives of people. There are many methods for detecting emotion via speech, different methods use different parameters of speech signal e.g. pitch, frequency, and many more. Emotion can be detected by differentiating between the values of various parameters. MFCC feature extraction is used for that purpose. These parameters play a vital role in detecting emotion from speech, changes in these parameters will result in a change in emotion. Along with that the system uses SVM to classify between the emotions.

Keywords: human interaction, pitch, MFCC, SVM, technology

Cite this Article: Kirti Shivaji Deshmukh, Sonal Gurudas Chavan, Aishwarya Deepak Bobhate, Sujata Bhairnallykar. Emotion Recognition based on Human Voice Tones.Current Trends in Information Technology. 2020; 10(2): 1–5p.