Wenzhounese Input Method and Language Technology

This project develops a digital input method for the Wenzhounese (溫州話) dialect, aiming to support the computational representation and preservation of an underrepresented Chinese language variety.

The system is implemented using the Rime input method framework, with a customized phonetic schema designed to represent the unique phonology of Wenzhounese.

Major components of the project include:

• Designing a phonetic transcription system for Wenzhounese
• Implementing a Rime-based input method schema
• Building dictionaries and mappings between phonetic input and Chinese characters
• Exploring language modeling approaches to improve typing efficiency
• Investigating the integration of large language models (LLMs) for dialect-aware language technologies

The broader goal of this work is to contribute to the digital preservation and computational accessibility of minority languages and dialects, enabling speakers to use Wenzhounese more easily in modern digital environments.