News » New Digital Humanities Project
HKUST Team Achieves World’s First Deep Learning Application for Preserving Indigenous Southeast Asian Languages
A research team led by Professor Giulio ONGARO, Assistant Professor of the Division of Humanities, has made a breakthrough in the study of an important indigenous language in Southeast Asia. Using cutting-edge technologies such as deep learning, the team has digitized materials related to the language for the first time and produced a dataset for further use in subsequent research. This study is supported by the Digital Humanities Seed Funding scheme of the Hong Kong University of Science and Technology.
Bridging the Digital Divide for Akha
The research team conducted in-depth research on Akha, a representative language. This language belongs to the Tibeto-Burman branch of the Sino-Tibetan language family and is widely used in China, Myanmar, Laos, and Thailand. Unlike the high-resource languages that deep learning typically focuses on, this language represents the rich cultural diversity of Southeast Asian indigenous peoples. The study systematically organized the materials of the Akha language for the first time and pioneered the way for research combining AI with indigenous languages.
From Archives to Digital Assets
The team systematically organized the existing academic research materials of Akha and combined visual deep learning models with AI technology to achieve the first digitization of the Akha-English dictionary. This work included the standardization of existing Akha archives, by converting several different Akha scripts into Common Akha Orthography (CAO). This will facilitate international collaboration in Akha studies and the preservation of Akha records in the AI era.
Based on this work, the team is working on developing digital teaching tools for Akha and further exploring the study of Akha audio samples. The team has developed a series of scripts to achieve automatic recognition, extraction, and conversion of Akha text. These tools will be used to preserve the rich cultural records of the Akha people and promote the public visibility of indigenous cultures in the AI era.
Interdisciplinary Innovation
The project involved three students from different majors across two local universities, fostering a unique cross-disciplinary exploration between Anthropology and Computer Engineering. This collaboration demonstrates how humanistic and cultural knowledge can serve as a primary driver for the frontier technology.
As the first project to preserve and research indigenous Southeast Asian languages based on AI technology, this study also provides a demonstration for similar indigenous language research in Asia, further proving the global leading position of the Hong Kong University of Science and Technology in the fields of anthropology, Asian languages and cultures, and digital humanities.
Text: Dr Steve Ma.
Layout & Design: Yifan Wang.