Issues with Caucasian Languages

[Plkmoi]

The model for Caucasian Languages are having very less accuracy. The languages of Abkhaz, Adyghe and Chechen have quite a lot of data in the internet and also bilingual corpuses. Chechen has a huge amount of Wikipedia data and Abkhaz, Kabardian, Lezgi, Adyghe, Ingush, Lak, Avar has Wikipedia data but the models using these languages have very less accuracy.

[jorgtied]

Contributions of clean aligned bilingual data sets would be very helpful to push the performance. Would you like to contribute? I welcome contributions to OPUS and that would feed into the NMT models as well ... Thanks!

[jorgtied]

More data would be the most practical thing that could help to improve the quality. I added an issue here to add some more data: Helsinki-NLP/OPUS-ingest#14
If you happen to know more sources of parallel data sets then, please, let me know. Thanks!

[AlexJonesNLP]

@Plkmoi It's possible that the Google Translate team is working on adding Caucasian languages in the somewhat near future. In fact, I somehow feel pretty good about that hunch. But I can't say any more for legal reasons at the moment (wink wink).

[AlexJonesNLP]

@dotsuzu I do have a contact! What's a good email so I can reach out and cc you?