English

Romanian Diacritics Restoration Using Recurrent Neural Networks

Computation and Language 2020-09-08 v1 Machine Learning

Abstract

Diacritics restoration is a mandatory step for adequately processing Romanian texts, and not a trivial one, as you generally need context in order to properly restore a character. Most previous methods which were experimented for Romanian restoration of diacritics do not use neural networks. Among those that do, there are no solutions specifically optimized for this particular language (i.e., they were generally designed to work on many different languages). Therefore we propose a novel neural architecture based on recurrent neural networks that can attend information at different levels of abstractions in order to restore diacritics.

Keywords

Cite

@article{arxiv.2009.02743,
  title  = {Romanian Diacritics Restoration Using Recurrent Neural Networks},
  author = {Stefan Ruseti and Teodor-Mihai Cotet and Mihai Dascalu},
  journal= {arXiv preprint arXiv:2009.02743},
  year   = {2020}
}

Comments

2 pages, 1 figure

R2 v1 2026-06-23T18:20:41.507Z