Tamyïz 🍉
Arabic Dialect Identification in Speech

Developed with ❤️🤍💚 by Badr al-Absi

This is a demo for the accurate and robust Transformer-based model for Spoken Arabic Dialect Identification (ADI). From just a short audio clip (5-10 seconds), the model can identify Modern Standard Arabic (MSA) as well as four major regional Arabic varieties: Egyptian Arabic, Peninsular Arabic (Gulf, Yemeni, and Iraqi), Levantine Arabic, and Maghrebi Arabic.

Simply upload an audio file 📤 or record yourself speaking 🎙️⏺️ to try out the model!