English-Ukrainian bidirectional neural machine translator, based on fastText word embeddings (sisg- model [1]) and default Transformer architecture [2] of the OpenNMT framework.
The following OPUS datasets [3] were used for training:
- WikiMatrix [4];
- XLEnt [5];
- Tatoeba [3];
- QED [6].
Launch translator:
Check out my article, related to this project.
- Bojanowski, P., Grave, E., Joulin, A., & Mikolov, T. (2017). Enriching word vectors with subword information. Transactions of the Association for Computational Linguistics, 5, 135-146.
- Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., ... & Polosukhin, I. (2017). Attention is all you need. arXiv preprint arXiv:1706.03762.
- Jörg Tiedemann, 2012, Parallel Data, Tools and Interfaces in OPUS. In Proceedings of the 8th International Conference on Language Resources and Evaluation (LREC'2012).
- Holger Schwenk, Vishrav Chaudhary, Shuo Sun, Hongyu Gong and Paco Guzman, WikiMatrix: Mining 135M Parallel Sentences in 1620 Language Pairs from Wikipedia, arXiv, July 11 2019.
- Ahmed El-Kishky, Adi Renduchintala, James Cross, Francisco Guzmán and Philipp Koehn, XLEnt: Mining Cross-lingual Entities with Lexical-Semantic-Phonetic Word Alignment, Online preprint, 2021.
- A. Abdelali, F. Guzman, H. Sajjad and S. Vogel, "The AMARA Corpus: Building parallel language resources for the educational domain", The Proceedings of the 9th International Conference on Language Resources and Evaluation (LREC'14). Reykjavik, Iceland, 2014. Pp. 1856-1862. Isbn. 978-2-9517408-8-4.