Question
#1
by
mrfakename
- opened
Hi,
Thanks for releasing this model! Looks super cool! Have you considered training a version on the official multilingual PL-BERT dataset, since it has longer context (the phoneme pairs are usually single sentences)?
Thanks!