Anyone else having this issue? If so, please add information to the Github issue if you’ve spotted any patterns.
Hi, in the Dutch (nl) language, I can’t confirm this behaviour. I had to validate the same text-clip a few times, but the recording was a different person. And this is not happening very often. (I don’t think my language is short on sentences and mulitple recordings of the same frase do not appear often).
note: for some reason I can’t yet comment on Github.
If you speed of the devil: I had a double today (tekst = Wij hebben hier te maken met een regelrechte aanslag op het subsidiariteitsbeginsel.)
I remember it quit well as a “difficult” word is part of it…
Same, I just had a duplicate in the Greek and in the Turkish datasets. I validated this clip yesterday and now it’s coming back:
“Γιατί κάτω από τα ρούχα τους δεν έχουν ούτε πουκάμισο!”
Also, I remember coming across this one in the past (Turkish):
“Yağmur mu yağıyor?”
…if it is of any importance. No problem validating them again.