A Portable System for Bidirectional Sign Language Translation for Deaf and Dumb

Sign language is a type of interaction used by individuals who are deaf or hard of hearing to interact with others. They have their very own syntax and vocabulary and are entirely natural. Although there are similarities between different sign languages, they are not global and are usually not mutually comprehensible. Because sign language is not a common language, figuring out what signs mean takes a lot of time and work. As a result, hearing-impaired people are rarely communicated with, isolating them from the rest of society. A mechanism for turning signs to text and text to signs is necessary to overcome this issue and ensure smooth communication. The existing systems produce good results when it comes to sign translation, yet there are occasional misclassifications. This can be enhanced even more by detecting and classifying the signs that are being misclassified. Our system provides the communication interface that performs the bidirectional sign language conversion and it uses Convolutional neural networks as the classification model which gave the training accuracy of 94.86%.