espnet

/

owsm_v3.1_ebf

like 13

Hi, thanks for the question! We haven't tested on many low-resource languages, so I'm not sure about the performance of that specific data. If that test data is ready, you could probably try running inference and check the result.

Thanks for responding! Which CV version is the model trained on? I think CV splits the data differently between versions, so I need to know the exact one.

We used this version: https://github.com/espnet/espnet/blob/1c1a12d2c7f613ac59cc996aad8777a8456e9e95/egs2/owsm_v3/s2t1/local/prepare_commonvoice.sh#L49

Thank you!