Sequence-to-sequence neural network models for transliteration

Abstract

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.

References in corpus (4)

• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Adaptive Computation Time for Recurrent Neural Networks
• Listen, Attend and Spell
• Sequence-to-Sequence Neural Net Models for Grapheme-to-Phoneme Conversion

Cited by in corpus (8)

• Arabic natural language processing: An overview
• Design Challenges in Named Entity Transliteration
• PheMT: A Phenomenon-wise Dataset for Machine Translation Robustness on User-Generated Contents
• Exact Hard Monotonic Attention for Character-Level Transduction
• ANETAC: Arabic Named Entity Transliteration and Classification Dataset
• Hard Non-Monotonic Attention for Character-Level Transduction
• RotLSTM: Rotating Memories in Recurrent Neural Networks
• Transliteration of Judeo-Arabic Texts into Arabic Script Using Recurrent Neural Networks