turjuman

<p align="center"> <br> <img src="https://github.com/UBC-NLP/turjuman/raw/master//images/turjuman_logo.png"/> <br> <p> <img src="https://github.com/UBC-NLP/turjuman/raw/master/images/turjuman.png" alt="AraT5" width="50%" height="50%" align="right"/>

Turjuman is a neural machine translation toolkit. It translates from 20 languages into Modern Standard Arabic (MSA). Turjuman is described in this paper: TURJUMAN: A Public Toolkit for Neural Arabic Machine Translation.

Turjuman exploits our AraT5 model. This endows Turjuman with a powerful ability to decode into Arabic. The toolkit offers the possibility of employing a number of diverse decoding methods, making it suited for acquiring paraphrases for the MSA translations as an added value.

Github: https://github.com/UBC-NLP/turjuman

Demo: https://demos.dlnlp.ai/turjuman

Paper: https://arxiv.org/abs/2206.03933

License

turjuman(-py) is Apache-2.0 licensed. The license applies to the pre-trained models as well.

Citation

If you use TURJUMAN toolkit or the pre-trained models for your scientific publication, or if you find the resources in this repository useful, please cite our paper as follows (to be updated):

@inproceedings{nagoudi-osact5-2022-turjuman,
  title={TURJUMAN: A Public Toolkit for Neural Arabic Machine Translation},
  author={Nagoudi, El Moatez Billah and Elmadany, AbdelRahim and Abdul-Mageed, Muhammad},
  booktitle = "Proceedings of the 5th Workshop on Open-Source Arabic Corpora and Processing Tools (OSACT5)",
  month = "June",
  year = "2022",
  address = "Marseille, France",
  publisher = "European Language Resource Association",
}