Amged-Salem/Ishara-

A graduation project for real-time translation of Egyptian Sign Language to text.” # Ishara: Real-time Egyptian Sign Language Translation System Ishara is the first application in Egypt addressing continuous Egyptian Sign Language translation, designed to bridge communication gaps for the deaf and hard-of-hearing community. This system translates Egyptian Sign Language to text and vice versa in real time, providing an essential tool for improving accessibility and inclusivity. The Ishara project leverages advanced machine learning techniques, computer vision, and natural language processing (NLP) to create a seamless translation system for continuous Egyptian Sign Language. Our mission is to enhance communication between sign language users and non-signers, making daily interactions more accessible. # Key Features Real-time Translation: Converts Egyptian Sign Language gestures into text and vice versa. Machine Learning: Powered by a custom-trained model to recognize continuous sign language gestures. Computer Vision & NLP: Detects hand gestures and facial expressions, and enhances text-to-sign accuracy. User-Friendly Interface: Intuitive and accessible interface designed for seamless communication. Community Collaboration: Developed with input from subject matter experts to ensure accuracy and cultural relevance. # Technologies Used Programming Languages: Python Machine Learning: TensorFlow, Keras Computer Vision: OpenCV Natural Language Processing: NLP libraries Other Tools: NumPy, pandas, scikit-learn # Data Collection As part of the Ishara project, our team collected a comprehensive dataset of sign language videos. These videos were used to train, test, and validate the machine learning models to ensure high translation accuracy. Use the video feed to capture continuous sign language gestures. View the real-time translation in the text box. # Contributing We welcome contributions to Ishara! Whether it's improving the model, refining the user interface, or adding new features, feel free to submit a pull request or open an issue.