|
International Journal of Computer Applications
Foundation of Computer Science (FCS), NY, USA
|
| Volume 187 - Issue 134 |
| Published: August 2026 |
| Authors: Mackenzie Rebelo, Vaishnavi Vasant Lotulkar, Vedant Gaonkar, Athary Prabhu, Nisha Costa, Manjusha Sanke, Prajacta Naik |
10.5120/ijcaf3e03846c333
|
Mackenzie Rebelo, Vaishnavi Vasant Lotulkar, Vedant Gaonkar, Athary Prabhu, Nisha Costa, Manjusha Sanke, Prajacta Naik . A Bidirectional Indian Sign Language Translation System using Speech Recognition and a 3D Avatar. International Journal of Computer Applications. 187, 134 (August 2026), 17-21. DOI=10.5120/ijcaf3e03846c333
@article{ 10.5120/ijcaf3e03846c333,
author = { Mackenzie Rebelo,Vaishnavi Vasant Lotulkar,Vedant Gaonkar,Athary Prabhu,Nisha Costa,Manjusha Sanke,Prajacta Naik },
title = { A Bidirectional Indian Sign Language Translation System using Speech Recognition and a 3D Avatar },
journal = { International Journal of Computer Applications },
year = { 2026 },
volume = { 187 },
number = { 134 },
pages = { 17-21 },
doi = { 10.5120/ijcaf3e03846c333 },
publisher = { Foundation of Computer Science (FCS), NY, USA }
}
%0 Journal Article
%D 2026
%A Mackenzie Rebelo
%A Vaishnavi Vasant Lotulkar
%A Vedant Gaonkar
%A Athary Prabhu
%A Nisha Costa
%A Manjusha Sanke
%A Prajacta Naik
%T A Bidirectional Indian Sign Language Translation System using Speech Recognition and a 3D Avatar%T
%J International Journal of Computer Applications
%V 187
%N 134
%P 17-21
%R 10.5120/ijcaf3e03846c333
%I Foundation of Computer Science (FCS), NY, USA
Communication is often challenging for hearing-impaired individuals due to the limited awareness of Indian Sign Language (ISL). Most existing solutions are text-based and lack an interactive and visual approach, making communication less effective and difficult to understand. This paper presents a bidirectional ISL translation system that converts speech or text into ISL using a 3D animated avatar and enables reverse translation from ISL to text or speech. By combining speech recognition, natural language processing, gesture recognition, and 3D animation, the proposed system provides an interactive approach to facilitate communication between hearing and hearing-impaired individuals. The trained sign recognition model achieved a test accuracy of 93.18% on the collected dataset, demonstrating its ability to recognize the sign classes used in the system.