Status: published
Communication between hearing-impaired individuals and normal users is often difficult because most people do not understand sign language. Sign language is one of the primary communication methods used by deaf and mute individuals, where hand gestures, finger movements, and facial expressions are used to express thoughts and emotions. This project presents a Sign Language Recognition System using Artificial Intelligence (AI) and Machine Learning (ML) techniques to reduce communication barriers and improve interaction in daily life. The proposed system uses a webcam or camera to capture hand gestures in real time and applies computer vision techniques for hand detection and feature extraction. OpenCV and MediaPipe are used to identify hand landmarks, finger positions, and gesture movements accurately. The extracted features are processed using deep learning models such as Convolutional Neural Networks (CNN) for static gesture recognition and Long Short-Term Memory (LSTM) networks for dynamic gesture recognition. The recognized gestures are converted into readable text or speech output, allowing effective communication between deaf individuals and normal users. The system is designed to be simple, cost-effective, accurate, and user-friendly, making it suitable for educational institutions, public services, healthcare environments, and assistive communication applications. The project also focuses on improving real-time performance, reducing recognition errors, and creating a scalable solution that can support future advancements such as multilingual sign language recognition and mobile-based applications.
Keywords: Sign Language Recognition (SLR), Artificial Intelligence (AI), Machine Learning (ML), Computer Vision, OpenCV, MediaPipe, Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM)