English

Sequence-to-sequence neural network models for transliteration

Computation and Language 2016-11-01 v1

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.

Keywords

Cite

@article{arxiv.1610.09565,
  title  = {Sequence-to-sequence neural network models for transliteration},
  author = {Mihaela Rosca and Thomas Breuel},
  journal= {arXiv preprint arXiv:1610.09565},
  year   = {2016}
}
R2 v1 2026-06-22T16:36:24.629Z