Vietnamese Sign Language Recognition for People with Hearing Impairment using Digital Library
Tran Thi Thuy Ha1, Phan Thi Ha2
1Dr. Tran Thi Thuy Ha, Lecturer, Faculty of Information Technology, Posts and Telecommunications Institute of Technology (PTIT) in Vietnam, Ha Dong, (Ha Noi), Vietnam.
2Phan Thi Ha, Lecturer, Faculty of Information Technology, Posts and Telecommunications Institute of Technology (PTIT) in Vietnam, Ha Dong, (Ha Noi), Vietnam.
Manuscript received on 20 June 2026 | First Revised Manuscript received on 27 June 2026 | Second Manuscript Accepted on 08 July 2026 | Manuscript Accepted on 15 July 2026 | Manuscript published on 30 July 2026 | PP: 7-11 | Volume-15 Issue-8, July 2026 | Retrieval Number: 100.1/ijitee.I128715090826 | DOI: 10.35940/ijitee.I1287.15080726
Open Access | Editorial and Publishing Policies | Cite | Zenodo | OJS | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This paper proposes a method for building a specialised sign language dataset for deaf-mute sign language recognition in digital libraries, comprising 120 signs and 65 phrases, collected from 25 people with hearing impairments in Hanoi and Ho Chi Minh City. The data were collected under various conditions and yielded approximately 125,000 video samples after processing. Based on experimental results with Random Forest, pure LSTM, and CNN-LSTM models, the CNN-LSTM model yields the best results and is selected for the development of Vietnamese sign language recognition services, demonstrating high performance, stability, and suitability for practical implementation requirements.
Keywords: Random Forest, Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), CNN-LSTM, Hearing Impairment, Identification, Sign Language, Library, Phrase.
Scope of the Article: Computer Science and Engineering