Human beings have a natural ability to see, listen and interact with their external world. Unfortunately, there are some people who are differently abled and do not have the ability to use their senses to the best extent possible. Such people depend on other means of communication like sign language. This presents a major roadblock for people in the hearing and speech impairment communities when they try to engage in interaction with others who are unfamiliar with the sign language, especially in their educational, social and professional environments. Therefore, it is necessary to have a gesture recognition or sign language detection system to bridge this communication gap. The main aim of the proposed system is to develop a smart glove for real-time gesture recognition using IoT. A mobile application is used as an interface to convert gestures into speech or voice output.