Hi all,
I know the docs say document translation picks one source language per file.
My problem is that my files don't really have one, a single .docx might have Dutch, French, and German paragraphs mixed together, and I need the whole thing in English.
Is there a way to handle this with the API?
I know I could fall back to the regular text translation endpoint and send things chunk by chunk with source_lang left blank, but that would explode the number of calls (and the cost) pretty fast. Is there a cleaner way to do this on the document side, or is splitting the only realistic path?
Doing it by hand isn't going to scale once there's any real volume.
Has anyone done this before?
Thanks.