Year: 2025
Type: article
Abstract: — Sign Language serves as the pri- mary medium of communication for millions of people who are hearing-impaired or speech- disabled. Despite its importance, the lack of un- derstanding of Sign Language among the general population creates a major communication bar- rier. To bridge this gap, we present a real-time Indian Sign Language Recognition (ISLR) sys- tem powered by deep learning and computer vi- sion techniques. Our proposed system employs Convolutional Neural Networks (CNN) combined with OpenCV for gesture detection and classifi- cation, enabling automatic recognition of hand signs from live video input. The recognized signs are then converted into meaningful text and speech output, allowing seamless communi- cation between hearing-impaired individuals and the larger community. This system is designed to be lightweight, scalable, and deployable on both desktop and mobile platforms. The primary con- tribution of this work lies in demonstrating the ef- fectiveness of CNN-based models for Indian Sign Language recognition, highlighting potential use cases in education, healthcare, and day-to-day communication. Furthermore, future integration with mobile applications and cloud services can make this solution widely accessible and impact- ful at a societal level.
DOI: 10.55041/ijsrem52686
Authors: Khyati Prajapati, Dr. Kapil Aggarwal, Parthesh Patel, Trupti Parmar
Institution: Parul University
Language: English
Subfield: Human-Computer Interaction
Field: Computer Science
Domain: Physical Sciences
SDG: Reduced inequalities
Open Access status: bronze
OpenAlex ID: W4414345311
FWCI:
Cites:
Cited by:
Related to: