Spaces:

---

omneity-labs

/

lid-benchmark

like 5

Running

Hello @boffire ! Sure, we can include it in the next update.

Do you have a paper or details on how language codes are mapped by your model, what kind of normalization is done etc..?

Hi @omarkamali ,
It is not mine : https://pleias.ai/blog/blogcommonlingua

It is from PleIAs @Pclanglais

Oh I see. Let me see what I can do!

CommonLingua is added to the benchmark. You can also check the individual results in the associated dataset:
https://huggingface.co/datasets/omneity-labs/lid-benchmark

Enjoy!