Sequence-to-sequence neural network models for transliteration
Published in
arXiv.org(2016)External Links:
TL
TL;DR
It is demonstrated that neural sequence-to-sequence models obtain state of the art or close to state-of-the- art results on existing datasets.
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.
References17 items
1
Adaptive Computation Time for Recurrent Neural Networks
2
TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
3
Sequence-to-sequence neural net models for grapheme-to-phoneme conversion
4
Grapheme-to-phoneme conversion using Long Short-Term Memory recurrent neural networks
5