Sequence-to-sequence neural network models for transliteration
This addresses transliteration for machine translation and software internationalization, but it is incremental as it applies an existing method to a new dataset.
The paper tackled transliteration by showing that neural sequence-to-sequence models achieve state-of-the-art or near state-of-the-art results on existing datasets, and they open-sourced a new Arabic to English dataset and trained models to improve accessibility.
Transliteration is a key component of machine translation systems and software internationalization. This paper demonstrates that neural sequence-to-sequence models obtain state of the art or close to state of the art results on existing datasets. In an effort to make machine transliteration accessible, we open source a new Arabic to English transliteration dataset and our trained models.