Danang Arbian Sulistyo, Aji Prasetya Wibawa, Wayan Firdaus Mahmudy, Fadhli Almu’Iini Ahda, Andrew Nafalski
This study addresses the critical need to preserve and revitalize the Javanese language, which despite its widespread popularity, faces challenges as a low-resource language in Indonesia. The decline in Javanese proficiency among younger generations poses a significant threat to the language's cultural significance and heritage. To address this issue, this study introduces an innovative approach to machine translation, focusing on the development of a robust Indonesian-Javanese translation system. Utilizing advanced neural machine translation (NMT) techniques, including Long Short-Term Memory (LSTM) networks, the proposed system aims to bridge the linguistic gap between Indonesian and Javanese. Special attention was given to the unique linguistic characteristics and challenges of Javanese, with the goal of achieving exceptional translation accuracy and fluency. Through extensive experimentation and evaluation, this study aims to demonstrate the effectiveness of the translation system in facilitating cross-cultural communication and language preservation efforts within the Javanese-speaking community. By emphasizing the significance of Javanese as a widely spoken yet under-resourced language, this study underscores the importance of innovative technological solutions in safeguarding linguistic diversity and cultural heritage. Through its contributions, the research seeks to address the pressing need for language preservation and revitalization, particularly in the context of low-resource languages like Javanese. © 2025, Ikatan Ahli Informatika Indonesia. All rights reserved.
Information Technology, Technology and Design Faculty, Institut Teknologi dan Bisnis Asia, Malang, Indonesia; Electrical Engineering and Computer Science, State University of Malang, Indonesia; Computer Science Department, Brawijaya University, Malang, Indonesia; Electrical Engineering, University of South Australia, Australia
