to save you a lookup: The Arabic text "رجمة نانسي قنقر" translates to English as: "Nancy Qanqar's translation" or "Translation by Nancy Qanqar" "رجمة" means "translation" and "نانسي قنقر" is the name "Nancy Qanqar"
And it seems to be because the training data is largely unofficial subtitles from movies. Which often have a string like "Translated by X" at the end of the movie which is often silent while credits roll.
Complete silence is always hallucinated as "ترجمة نانسي قنقر" in Arabic
11–20 of 354 posts
Re: Complete silence is always hallucinated as "ترجمة نانسي قنقر" in Arabic
#12to save you a lookup: The Arabic text "رجمة نانسي قنقر" translates to English as: "Nancy Qanqar's translation" or "Translation by Nancy Qanqar" "رجمة" means "translation" and "نانسي قنقر" is the name "Nancy Qanqar"
And it seems to be because the training data is largely unofficial subtitles from movies. Which often have a string like "Translated by X" at the end of the movie which is often silent while credits roll.
Re: Complete silence is always hallucinated as "ترجمة نانسي قنقر" in Arabic
#13Re: Complete silence is always hallucinated as "ترجمة نانسي قنقر" in Arabic
#14Earlier quoted context omitted.
And it seems to be because the training data is largely unofficial subtitles from movies. Which often have a string like "Translated by X" at the end of the movie which is often silent while credits roll.
Looks like they used more official sources for German - there, silence is apparently hallucinated as "Untertitelung des ZDF für funk, 2017" according to one of the comments on the issue. Which makes sense, as the public broadcasters' "Mediathek" is probably the largest freely available resource of subtitled videos in Germany. I wonder if the ZDF gave its approval for it being used for LLM training though?
I am pretty sure they didn't get asked.
Re: Complete silence is always hallucinated as "ترجمة نانسي قنقر" in Arabic
#15Re: Complete silence is always hallucinated as "ترجمة نانسي قنقر" in Arabic
#16Earlier quoted context omitted.
And it seems to be because the training data is largely unofficial subtitles from movies. Which often have a string like "Translated by X" at the end of the movie which is often silent while credits roll.
Looks like they used more official sources for German - there, silence is apparently hallucinated as "Untertitelung des ZDF für funk, 2017" according to one of the comments on the issue. Which makes sense, as the public broadcasters' "Mediathek" is probably the largest freely available resource of subtitled videos in Germany. I wonder if the ZDF gave its approval for it being used for LLM training though?
Re: Complete silence is always hallucinated as "ترجمة نانسي قنقر" in Arabic
#17Garbage in, garbage out. If the training dataset (accidentally) paired silence (`X_train`) with `رجمة نانسي قنقر` tokens (`y_pred`), then any silence will always be translated to that. Fortunately, this particular problem is easy to fix--just detect and remove silent parts before API call. This also has a side benefit of saving you money on transcription.
Re: Complete silence is always hallucinated as "ترجمة نانسي قنقر" in Arabic
#18In Italian as well there are random hallucination when parsing silence, something like: “Thank you for watching”, “Subtitles by…”
Re: Complete silence is always hallucinated as "ترجمة نانسي قنقر" in Arabic
#19Way to go Nancy! Keep up the good work, ya crazy bastard!