Sistem Pendeteksi Bahasa Isyarat SIBI Menggunakan LSTM Berbasis OpenCV dan MediaPip

Read full text Summarise with VIRAT
Authors: Ahmad Sarjianto, Melladia Melladia
Cited by: 0
References: 11

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This study aims to develop a Sign Language Translation System which is specifically for the Indonesian Sign Language System (SIBI) based on artificial intelligence (AI) and computer vision which aims to help communication between deaf/mute people and the general public. using the Long Short-Term Memory (LSTM) method, taking important data from Sign Language hand movements and combined with OpenCV and MediaPip. This system is designed with a web-based interface that will display translations in text form in real-time. The testing was conducted on a dataset consisting of SIBI alphabets and basic words, with the highest accuracy reaching 0.85 or 85% for basic words, and 0.45 or 45% for alphabet recognition.In conclusion, this research produced a system capable of automatically translating sign language by utilizing web technology for the interface, and OpenCV, MediaPip, and Long Short-Term Memory (LSTM) for the translation process.This system has great potential to reduce communication barriers between the general public and individuals with hearing or speech impairments, although further development is required to improve its accuracy.

Fields of Study & Subjects

Hand Gesture Recognition Systems Human Pose and Action Recognition Hearing Impairment and Communication

Publication details

Journal / source
JURNAL ILMIAH INFORMATIKA
Volume
14
Issue
01
Pages
8–14
Publication date
31 March 2026
Language
EN