American Sign Language Recognition Using GRU and LSTM

S. Pooja, Prem Sai N, Aravinth A, Prithiviraj R, Manikandan P

2025

Abstract

People with hearing loss face many challenges when it comes to communication since they frequently lack the tools needed to interact with others in a meaningful way. Although sign language is an essential tool for communication, its automated recognition is challenging since motions are dynamic in nature. This paper presents a Sign Language Recognition model that uses networks of Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU) to recognize hand gestures for the American Sign Language (ASL) Alphabet. Because GRU and LSTM effectively capture both the temporal and spatial aspects of hand gestures, the model is well-suited to handle the sequential nature of ASL movements. Critical features are extracted by the model after preprocessing input data, such as video frames or skeletal hand tracking data. The GRU and LSTM networks receive these features and use them to learn the time-dependent patterns of hand movements in order to correctly classify the corresponding ASL letters. The accuracy of the system is evaluated in real-time scenarios after it has been trained on a labeled dataset. This method facilitates smoother interactions and improves communication for people with hearing loss by offering real-time ASL identification. The model does a good job of identifying hand movements, but it has problems with computational complexity, especially when used on devices with little processing power. But compared to conventional models, the recognition process is more effective with the combination of GRU and LSTM networks, which makes this system a potential step toward helping people with hearing loss communicate.

Paper Citation

in Harvard Style

Pooja S., N P., A A., R P. and P M. (2025). American Sign Language Recognition Using GRU and LSTM. In Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT; ISBN 978-989-758-763-4, SciTePress, pages 659-666. DOI: 10.5220/0013639700004664

in Bibtex Style

@conference{incoft25,
author={S. Pooja and Prem Sai N and Aravinth A and Prithiviraj R and Manikandan P},
title={American Sign Language Recognition Using GRU and LSTM},
booktitle={Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT},
year={2025},
pages={659-666},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013639700004664},
isbn={978-989-758-763-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 3rd International Conference on Futuristic Technology - Volume 3: INCOFT
TI - American Sign Language Recognition Using GRU and LSTM
SN - 978-989-758-763-4
AU - Pooja S.
AU - N P.
AU - A A.
AU - R P.
AU - P M.
PY - 2025
SP - 659
EP - 666
DO - 10.5220/0013639700004664
PB - SciTePress