Enhancing Infants Early Communication through Sign Language Recognition System using Raspberry Pi and Convolutional Neural Network

Early communication among infants is very helpful especially on the growth and mental ability of the child. Infants' early communication, as they have limited verbal skills and may struggle to communicate their needs and feelings effectively to their caregivers. While sign language has been shown to be an effective way to bridge this communication gap, caregivers may not have the necessary skills or knowledge to understand or teach signs to their infants. To address this problem, this research aims to develop and evaluate an automated sign language recognition system that can enhance early communication between infants and caregivers, thus improving infants' overall language development and quality of life. Using the internet of things and the models of machine learning algorithms for early recognition of sign language gesture results better understanding of mental acquisition of infants. This research explores the feasibility of using an Internet of Things (IoT) device, specifically a Raspberry Pi, combined with deep learning techniques to recognize baby sign language. The study uses a deep learning model to recognize sign language gestures and evaluates the impact of various variables, such as lighting conditions and background, on the accuracy of the model. The results demonstrate that the system is capable of accurately recognizing baby sign language gestures with an overall accuracy of 94.6%. This research has important implications for early intervention for hearing-impaired infants, as it allows for real-time monitoring and tracking of a baby’s development of sign language skills. The study recommends further exploration of using IoT devices and deep learning techniques for recognizing and monitoring baby sign language, as well as potential applications for assistive technologies and other areas of human communication.