This is a GPT2 model of Italian regional languages trained on collections of Italian "dialect poetry" by Luigi Bonaffini.
This is a multilingual model. Italians use the word "dialect" to describe their regional languages, but they are separate languages. And there's a lot of English in this dataset too.
The challenge of this project is to train a model to write the languages of Italy.
For those who do not know Italian, here's some (lowercase) text that you can type into the API box:
- oggi si parla il dialetto
- la sua poesia viene di
- ma non sempre trova
- Downloads last month