Need assistance / ideas - Translating large Japanese scanned PDF files

Akagi

Akagi

Akagi

Caporegime
Joined
29 Jul 2011
Posts
37,064
Location
In acme's chair.
Hi all,

I have several large PDF files which are in Japanese, I need to translate them to English with text-in-place so that the formatting and imagery are retained.

Google Lens does an excellent job of translating any and every part of these documents, however Google Docs Translate claims that it cannot translate scanned documents... Which makes no sense because I assume they would be using the same technology, but never mind...

I had the idea to try using a Japanese optical character recognition (OCR) software and then using Google Translate afterwards, however these have proven to be extremely unreliable and ruin the document formatting...

There are around 1500 pages, so while I could screenshot every single page and use Google Lens on my phone to translate them, I'm sure you can see why that really isn't an option! :p

Any ideas?

Thanks :)
 
Last edited:
Update: I can easily convert the PDF file to images, and then convert images back to a PDF file.
So services that can only translate images are fine, however Google Translate only does one image at a time.

I did find one that bulk translates images using Google translate, but it costs $100 for 250 un-watermarked images! How they think they can charge that for a script that automates a free service is just baffling.
I am now looking for a free / much cheaper one...
 
Last edited:
NVP said:
Write a script to parse them through Google Translate yourself?
Click to expand...

I would if I knew how. :p

I've since done about 750 pages manually :o

At a rate of one every 3 or 4 seconds its not too painful if I have some music going...
 
You must log in or register to reply here.
Back
Top Bottom