SHAMI-MT: A Syrian Arabic Dialect to Modern Standard Arabic Bidirectional Machine Translation System
Serry Sibaee, Omer Nacar, Yasser Al-Habashi, Adel Ammar, Wadii Boulila
2025-08-05
Summary
This paper talks about SHAMI-MT, a machine translation system that can translate back and forth between the Syrian Arabic dialect and Modern Standard Arabic using a special language model.
What's the problem?
The problem is that Arabic has many dialects that differ a lot from the formal Modern Standard Arabic, making it hard for machines to accurately translate between the dialect and the standard language.
What's the solution?
SHAMI-MT solves this by using the AraT5v2-base-1024 language model architecture designed to handle both dialect and standard forms, allowing the system to provide high-quality, two-way translations between Syrian Arabic and Modern Standard Arabic.
Why it matters?
This matters because it helps improve communication and understanding across different Arabic-speaking communities and supports language technologies that can work effectively with dialects and formal language.
Abstract
A bidirectional machine translation system, SHAMI-MT, bridges the gap between Modern Standard Arabic and the Syrian dialect using AraT5v2-base-1024 architecture, achieving high-quality translations.