A Smart Sign Language Interpreter for Medical Environments Using Deep Learning: Morocco Case Study

Abstract

This paper presents a smart system that aims to improve communication for individuals who primarily use sign language and those who do not master it. Innovative solutions in gesture recognition can effectively address and bridge the communication gap that exists between hearing individuals and people who are deaf, deafened, hard of hearing, or non-verbal. The proposed system integrates surface electromyography (EMG) and Inertial Measurement Unit (IMU) sensors with Convolutional Neural Networks (CNN) to accurately interpret Moroccan Sign Language (MSL) and convert it into spoken language. The system of interpretation of sign language holds the overarching goal of contributing to the social integration of individuals with disabilities in medical and hospital environments.