UTD-CRSS Submission for MGB-3 Arabic Dialect Identification: Front-end and Back-end Advancements on Broadcast Speech

TL;DR

This paper introduces advanced speech processing systems for Arabic dialect identification, combining various feature extraction techniques and classifiers, achieving state-of-the-art accuracy in the MGB-3 challenge.

Contribution

It presents novel combinations of front-end features and back-end classifiers, including GANs, for improved Arabic dialect identification performance.

Findings

Achieved 79.76% accuracy on test data.

Compared different feature extraction methods and classifiers.

Set new benchmark for Arabic dialect identification in MGB-3.

Abstract

This study presents systems submitted by the University of Texas at Dallas, Center for Robust Speech Systems (UTD-CRSS) to the MGB-3 Arabic Dialect Identification (ADI) subtask. This task is defined to discriminate between five dialects of Arabic, including Egyptian, Gulf, Levantine, North African, and Modern Standard Arabic. We develop multiple single systems with different front-end representations and back-end classifiers. At the front-end level, feature extraction methods such as Mel-frequency cepstral coefficients (MFCCs) and two types of bottleneck features (BNF) are studied for an i-Vector framework. As for the back-end level, Gaussian back-end (GB), and Generative Adversarial Networks (GANs) classifiers are applied alternately. The best submission (contrastive) is achieved for the ADI subtask with an accuracy of 76.94% by augmenting the randomly chosen part of the development dataset. Further, with a post evaluation correction in the submitted system, final accuracy is increased to 79.76%, which represents the best performance achieved so far for the challenge on the test dataset.