Sign Language Recognition And Translation Into Text And Speech

This project aims to allow easy and low- cost communication between people with hearing/speech disabilities and people without these disabilities, by introducing an IOT device medium. We propose a simple system that converts images to text which in turn convert speech. Images are acquired by a low-cost camera, these images are processed and used for training a neural network model. This trained model then converts the image inputs into suitable textual outputs. Once the text outputs are displayed, we can then convert them into speech. The IoT medium which will process all these tasks and make our project feasible is a raspberry-pi 3. The raspberry- pi is a credit card-sized single-board micro- computer chip that is powerful enough to run neural networks and image processing. This microcomputer board also has general purpose I/O (GPIO) pins, allowing us to attach hardware components such as a camera, display, and speaker. The finished product is meant to be affordable and efficient.