Development of a smart system to translate sign language for deaf and mute individuals

Abstract

Communication is another major barrier that the deaf and the mutes are faced with in mostly a vocal society. More than 13 million individuals in Bangladesh experience some hearing impairments and impediments on speech. They use sign language thus they cannot communicate well most of the time because the one being signed can not understand sign language. The idea of the proposed project is to design a smart wearable device that can be used to translate sign language into speech by hard of hearing persons and actors with impaired voices. The system involves flex sensors, a 6 axis accelerometer-gyro which is mounted into a glove in order to pick up the hand motions and hand postures. The data is processed in the form of sensor data transmitted to a microprocessing unit that has a machine learning ready with such information. In case such a person is a deaf mute, he/she will move his/her hands and our model will identify it and provide us with the corresponding sign (word) by means of a speaker. This solution addresses the shortfalls of the earlier systems since it allows translation in real time regardless of the environmental situation. The solution also improves the accessibility to communication, social inclusion, and safety among the users especially in emergencies and the day-to-day situations.