Language distribution of training data

[rsmlgen]

Hi,

May I ask how is the language distribution of the data that is used to train this model?
It says Emilia-50k is used but I could not find such subset of Emilia provided. Is it expected to be similar distribution with Emilia-101k?

[ZhengRachel]

Same question here. Are languages other than English also included in the Emilia-50k dataset? I am curious about FACodec's generalization on out-of-domain languages.