Hand Speak’s - Sign Language Recognition System

Abstract

HAND SPEAK’S is a sign language recognition system. The system is designed to recognize and interpret a wide range of sign language gestures, converting them into readable text or spoken words in real-time. The sign language recognition system presented in this paper represents a significant step towards improving accessibility and inclusivity for the deaf and hard-of-hearing communities. The system integrates a user-friendly interface that allows users to interact with the recognition software seamlessly. This research paper presents a novel approach to recognize American Sign Language (ASL) at the sentence level utilizing Convolutional Neural Networks (CNN) with TensorFlow and OpenCV. In this paper, we explore the challenges associated with ASL recognition, including the complexities of hand gestures, spatial variations, and real-time processing requirements. The proposed method aims to enhance the accuracy and efficiency of ASL recognition systems, thereby facilitating smoother interaction and understanding between individuals using sign language and those who do not. The study discusses the dataset preparation, model architecture, training process, and evaluation metrics. Experimental results demonstrate the effectiveness and robustness of the proposed approach in recognizing ASL sentences accurately. Our method provides 97.2 % accuracy. Future work will focus on expanding the enhancing real-time performance, and exploring multilingual sign language recognition capabilities.