Human-machine interaction has seen a rise in the use of emotion recognition systems (ERS), which use internet of things (IoT) and artificial intelligence (AI). Automated emotion recognition (AEE) is essential for industries like technology, marketing, and human-robot interaction where human emotional responses operate as signals. Machines can recognize and analyze human emotions thanks to recent developments in sensors and information technologies. Analyzing behavior, facial expressions, physiological signs, voices that are gathered using various sensors are necessary for emotion identification. The papers that are considered review various techniques for detecting emotions. The benefits and drawbacks of multiple sensors are also compared. The aim of this work is to help researchers to select optimum sensor and technique when developing emotion recognition system.