Real-Time American Sign Language Realization Using Transfer Learning With VGG Architecture

Sign language is a barrier between the hearing impaired and the hearing community. In the modern world, despite the technical advancements, it is still a struggle for specially-abled people to communicate and express their thoughts with the rest of the world. A system is built that will reduce the gap for the specially-abled, the system recognizes the hand gesture and translates it into corresponding letters. ASL data set is used. The software is built on Python and inherits the power of deep learning. A convolutional neural network is used for image classification. The system was able to classify all the alphabets from A-Z with an accuracy of 98.89 percent. The software aims to eradicate the communication barrier between the hearing-impaired community and the rest of the world. The software provides a medium for the inarticulate community to express their thoughts to everyone out there.