Real-time Arabic sign language translator Using media pipe and LSTM
- Communication and Computers Engineering Department Alshorouck Academy, Cairo, Egypt
- Communication and Computers Engineering Department Alshorouck Academy, Cairo, Egypt
- Communication and Computers Engineering Department Alshorouck Academy, Cairo, Egypt
- Department of Communication and Computer Engineering, Higher Institute of Engineering, El Shorouk Academy, Cairo 11837, Egypt.
- Computer Engineering Department, College of Computer Science and Engineering, Hail University, Hail, Saudi Arabia.
* Author to whom correspondence should be addressed.
PLOMSE AI 2024, 4(1)
- Submission received
- 7 May 2024
- Accepted
- 7 May 2024
- Published
- 7 May 2024
K. Hussin, S., Mohamed, O., Mohamed, M., Ahmed, E. & Mahmoud, O., (2024) “Real-time Arabic sign language translator Using media pipe and LSTM”, PLOMSE AI 4(1).
Abstract
People who are dumb and deaf have difficulty communicating daily. Artificial intelligence (AI) developments have allowed the breaking down of this communication barrier. As a result of this work, an Arabic sign language (ASLT) letter recognition system has been created. The ASLT recognition system employs BILSTM with a media pipe structure to interpret depth data and enhance the social interaction of hearing-impaired people. Depending on user input, the suggested approach would automatically detect and identify Arabic and hand-sign alphabet letters. The proposed model should have a 98% accuracy rate in identifying ASLT for letters, 96% for words, and 100% for digits from 0 to 9. We conducted comparative research to evaluate our method, and the results showed that it is more accurate at differentiating between static signs than earlier studies using the same dataset.
Keywords
© 2024 by the authors. Licensee PLOMSE, Richardson, TX, USA. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution 4.0 (CC BY 4.0) license.