AI-Based Sign Language Recognition and Translation System

DOI

https://www.doi.org/10.59256/ijsreat.202606060009

Published In

Volume 6, Issue 6 (2026)

Published Online

November-December 2026

Pages

56-61

Article Type

Original Article

Check for updates

The AI-Based Sign Language Recognition and Translation System is a computer vision-based application designed to facilitate communication between individuals with hearing and speech impairments and people who do not understand sign language. The system captures hand gestures through a webcam and uses OpenCV and MediaPipe to extract hand landmarks for gesture recognition. A machine learning model processes the extracted features to identify corresponding English alphabet signs and display the recognized letters as text through a Streamlit-based user interface. The system supports both single-hand and two-hand gesture recognition based on the trained datasets. Text-to-speech technology is used to convert recognized text into audible speech, while speech recognition enables spoken input to be converted into text for reverse communication. The system also incorporates multilingual translation and an Emergency Mode to support communication in different situations. Python serves as the primary programming language, while OpenCV, MediaPipe, NumPy, and Joblib support image processing, landmark extraction, and model prediction. The proposed system aims to reduce communication barriers and provide an accessible platform for sign language-based interaction. However, recognition performance depends on lighting conditions, camera quality, hand positioning, and variations in gestures. Further testing with diverse users and real-world conditions is required to evaluate the system's generalization and reliability.