ADI-20: Arabic Dialect Identification dataset and models

TL;DR

This paper introduces ADI-20, a comprehensive Arabic dialect dataset with 3,556 hours covering all dialects and MSA, and evaluates state-of-the-art models for dialect identification, analyzing data size and model complexity effects.

Contribution

The paper provides a new extensive Arabic dialect dataset, ADI-20, and evaluates multiple models, including fine-tuned ECAPA-TDNN and Whisper-based systems, for dialect identification.

Findings

Small decrease in F1 score with 30% training data

Model performance affected by data size and parameter count

Open-sourced dataset and models for future research

Abstract

We present ADI-20, an extension of the previously published ADI-17 Arabic Dialect Identification (ADI) dataset. ADI-20 covers all Arabic-speaking countries' dialects. It comprises 3,556 hours from 19 Arabic dialects in addition to Modern Standard Arabic (MSA). We used this dataset to train and evaluate various state-of-the-art ADI systems. We explored fine-tuning pre-trained ECAPA-TDNN-based models, as well as Whisper encoder blocks coupled with an attention pooling layer and a classification dense layer. We investigated the effect of (i) training data size and (ii) the model's number of parameters on identification performance. Our results show a small decrease in F1 score while using only 30% of the original training data. We open-source our collected data and trained models to enable the reproduction of our work, as well as support further research in ADI.