Translating Dialects with Natural Language Processing: A Case Study of Hokkien

Speech-to-speech translation technologies today hold great promise for overcoming linguistic barriers. However, their application to low-resourced languages remains largely unexplored in practical terms. One such language is Hokkien, a Southern Min dialect spoken by millions in Taiwan and Southeast Asia. Hokkien’s tonal system, phonetic diversity, and lack of standardised linguistic resources present significant challenges for its integration into modern language technologies. This project addresses these challenges by developing a speech-to-speech translation system specifically for Hokkien.