You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Language ID for multilingual spans on multilingual pages is not supported by Chrome right now. CLD3 will return the most prevalent language it finds. Looking at your dump, there is mostly English text with some short Spanish text segments, so the model returns English. We don't do any special processing for Twitter.
This is found in crbug.com/809243.
Repro twitter page: https://twitter.com/paurubio
Almost all tweets are Spanish, but it's still identified as English when Chrome tries to detect page language using CLD.
I'm also attaching Chrome's text dump that's passed to CLD for language detection: dump.txt
The text was updated successfully, but these errors were encountered: