Automatic Arabic diacritization is useful in many applications, ranging from reading support for language learners to accurate pronunciation predictor for downstream tasks like speech synthesis. While most of the previous works focused on models that operate on raw non-diacritized text, production systems can gain accuracy by first letting humans partly annotate ambiguous words. In this paper, we propose 2SDiac, a multi-source model that can effectively support optional diacritics in input to inform all predictions. We also introduce Guided Learning, a training scheme to leverage given diacritics in input with different levels of random masking. We show that the provided hints during test affect more output positions than those annotated. Moreover, experiments on two common benchmarks show that our approach i) greatly outperforms the baseline also when evaluated on non-diacritized text; and ii) achieves state-of-the-art results while reducing the parameter count by over 60%.

I love machine learning, and deep learning as well. Yet, from day one of learning both, I discovered that to be better in this, I must be good at reading research papers. It takes a lot of training to learn how to read papers and summarize them. So I decided to share the world with some of those moments of decoding a research paper into summarized English text. Today I will cover Arabic Text Diacritization Using Deep Neural Networks.