Indonesian Sign Language Recognition

The Indonesian Sign Language Recognition (BISINDO) application is a tool designed to translate Indonesian sign language gestures into text or speech. This application is developed using Python along with key libraries such as Mediapipe for real-time hand tracking and gesture recognition, Keras for building and training deep learning models, and Flask for creating the web-based interface.

BISINDO allows users to interact with the application using sign language gestures captured by a camera. The Mediapipe library processes these gestures in real-time, identifying specific hand movements and positions. The recognized gestures are then passed through a Keras-based deep learning model, which interprets the gestures and translates them into corresponding text or audio representations.

Flask is used to create the user interface, enabling users to access the application through a web browser. Users can perform sign language gestures in front of their device's camera, and the application will promptly provide the translated output, whether as text displayed on the screen or as spoken words using synthesized speech.

The combination of Python, Mediapipe, Keras, and Flask forms a powerful framework for building a sign language recognition and translation system. This application aims to enhance communication accessibility for individuals who use Indonesian sign language by providing a technology-assisted method of understanding and interpreting sign language gestures.