With You - Indian Sign Language Detection and Alert System
Shreya Daga, Atharva Dusane, Dyuti Bobby
Abstract
The global deaf, non-speaking, and hard-of-hearing community relies heavily on sign language to facilitate communication. This paper will delve into the development of a detection system for the Indian Sign Language (ISL) that puts convenience and inclusivity first for this group. To accomplish this goal, the system uses a Long Short-Term Memory (LSTM) artificial neural network. Currently, the system can recognize 6 dynamic signs, 6 emergency signs, and 26 static ISL alphabets. The emergency signals use certain non-verbal cues to automatically dial the designated emergency services number. The emergency's type, the user's IP address, and their location are all included in the message that is delivered. By enabling communication through ISL, this system has been created to provide accessibility for the community of people with hearing disabilities. The artificial neural network enables the system to acquire the capability of learning and identifying various signals. Additionally, the use of emergency indicators can be extremely important in circumstances when verbal communication is impossible as the system can immediately notify emergency services and give them the information they need, to deliver quick aid. Overall, the hearing-impaired people in India may benefit greatly from this system's contributions to their safety and well-being.