Finetune the model on other writing systems like Arabic or Hebrew
Hi!
The model works astonishingly well on Nordic data. Very good job!
The model was trained primarily on handwritten text that uses basic Latin characters (A-Z, a-z) and includes Nordic special characters (å, ä, ö). It has not been trained on non-Latin alphabets, such as Chinese characters, Cyrillic script, or other writing systems like Arabic or Hebrew
Did you try to fine tune such a model on other writing systems like Arabic or Hebrew? I'm trying to do that but no such successfull attempts anywhere that I could find.
Hello!
Thanks for your kind words!
Unfortunately, our interests are mainly in Finnish and Swedish languages at the moment. We have tried finetuning on Cyrillic script, but that is still a work in progress. In the case of Arabic and Hebrew texts, we have very limited material on those scripts, so training our own model is not in our interests:(. Nevertheless, I hope you find what you are looking for!
Best regards,
AI team from National Archives of Finland