Neural Arabic Text Diacritization: State of the Art Results and a Novel Approach for Machine Translation

TL;DR

This paper introduces advanced deep learning models for Arabic text diacritization, demonstrating improved performance and proposing a novel method to enhance machine translation using diacritized text.

Contribution

The paper presents new neural network models for Arabic diacritization that outperform existing methods and introduces the Translation over Diacritization (ToD) approach for improved machine translation.

Findings

Models outperform or match existing methods without language-dependent post-processing

Diacritics improve NLP task performance, especially machine translation

Proposed models are tested on a standard benchmark dataset

Abstract

In this work, we present several deep learning models for the automatic diacritization of Arabic text. Our models are built using two main approaches, viz. Feed-Forward Neural Network (FFNN) and Recurrent Neural Network (RNN), with several enhancements such as 100-hot encoding, embeddings, Conditional Random Field (CRF) and Block-Normalized Gradient (BNG). The models are tested on the only freely available benchmark dataset and the results show that our models are either better or on par with other models, which require language-dependent post-processing steps, unlike ours. Moreover, we show that diacritics in Arabic can be used to enhance the models of NLP tasks such as Machine Translation (MT) by proposing the Translation over Diacritization (ToD) approach.