This article is an overview of a practical implementation of monitoring the emotional state of a student with special needs. In this article, we present an in-depth review of current technologies for detecting human emotions. The purpose of this work is to create an AI system to help teachers and psychologists to observe children with special needs while they are learning or passing tests. The object is the methods and tools for emotion recognition. This system is designed to solve the problems of recognizing the emotional state of the student and helping him to properly respond to the emotional manifestations of other people. The subject of this system are processes of activity of this AI. We conclude the investigation by highlighting the aspects that require further research and development.