Translating foreign language books - Tips & Tricks

jDooz

Member
Joined
Oct 19, 2022
Messages
17
Reaction score
52
I am definitely interested, please share


I have been trying to translate a book for awhile now, over at Juden-Elend im Lande der Romanows, Paul Dimidow (1891). I managed to translate the first chapter, but it took me more time than I want to admit. Any suggestions?
Try to find source material that is either in epub, docx, mobi, htm(l), azw3, or fb2. That way you can just convert it to docx, toss it into Google translate, and fix the few remaining translation errors, thereby retaining the original chapters, Footnotes, etc, all while not having the translation messed up due to page breaks, section breaks, hyphenations, page numbers, headers, Footnotes, etc. that will invariably result from using a pdf as your source. If you cannot find your text rendered into any of the above mentioned formats, you'll have to toss your PDF into ABBYY FineReader. Here's a torrent to the full, portable version (which conviently supports old fonts in multiple languages)!

If you don't already have it, download & install uTorrent 2.2.1 Build 25302
Then download the following torrent and click on the portable installer.
ABBYY FineReader PDF 15.0.114.4683 (18.08.2022) (2021) PC | RePack & Portable.
That program should help you tremendously in cases where obtaining a PDF or, God forbid, a poorly scanned PDF of a foreign, handmade book, set with old font type, is the only option available.

I've been doing this kind of thing for nearly a decade now. Let me know if you need any further assistance. I'm always willing to pass down all of the various tricks and tips I picked up along the way as I never had a mentor who could've prevented me from wasting countless hours of inefficiency and wasted time when I first started doing this lol

I can help you do things like remove various DRMs, access public & invite-only ebook websites, utilize handy macros for use in Microsoft Word, etc.

Good luck with your project!
 
Try to find source material that is either in epub, docx, mobi, htm(l), azw3, or fb2. That way you can just convert it to docx, toss it into Google translate, and fix the few remaining translation errors, thereby retaining the original chapters, Footnotes, etc, all while not having the translation messed up due to page breaks, section breaks, hyphenations, page numbers, headers, Footnotes, etc. that will invariably result from using a pdf as your source. If you cannot find your text rendered into any of the above mentioned formats, you'll have to toss your PDF into ABBYY FineReader. Here's a torrent to the full, portable version (which conviently supports old fonts in multiple languages)!

If you don't already have it, download & install uTorrent 2.2.1 Build 25302
Then download the following torrent and click on the portable installer.
ABBYY FineReader PDF 15.0.114.4683 (18.08.2022) (2021) PC | RePack & Portable.
That program should help you tremendously in cases where obtaining a PDF or, God forbid, a poorly scanned PDF of a foreign, handmade book, set with old font type, is the only option available.

I've been doing this kind of thing for nearly a decade now. Let me know if you need any further assistance. I'm always willing to pass down all of the various tricks and tips I picked up along the way as I never had a mentor who could've prevented me from wasting countless hours of inefficiency and wasted time when I first started doing this lol

I can help you do things like remove various DRMs, access public & invite-only ebook websites, utilize handy macros for use in Microsoft Word, etc.

Good luck with your project!
This looks like a big help. Haven't involved with Torrents installations; but Google translate has the 5K character limit which inhibits ...
 
This looks like a big help. Haven't involved with Torrents installations; but Google translate has the 5K character limit which inhibits ...
Not if you upload a word document into Google translate (.docx), as mentioned above. Now although there is no character limit, the file does still have to be under 10 MB. However, what you can do if your document is over 10MB is convert your document to an ePub, then compress said ePub, before finally converting ePub back into docx.

Also, Torrents are super easy. Just install uTorrent, then click on the ABBYY torrent link and it should automatically open it with uTorrent by default. Just choose which directory you want to download your torrent and click OK. Wait until it's finished downloading, then open the folder where you downloaded it and click on the portable installer. It should install everything you need in that exact folder.
I am definitely interested, please share
Forgot to respond to this part of your comment. I'm super busy at work but maybe this weekend, I'll try to compile all of my translations of Nosovsky/Fomenko's How it was in Reality Series (save for Русь-Орда в фундаменте Европы и Византии [Russia-Horde in the Foundation of Europe and Byzantium], as they've only released a PDF of that book, no ePub, fb2, etc.), none of which have ever been released in English, as far as I'm aware.
 
Last edited:
Try to find source material that is either in epub, docx, mobi, htm(l), azw3, or fb2. That way you can just convert it to docx, toss it into Google translate, and fix the few remaining translation errors, thereby retaining the original chapters, Footnotes, etc, all while not having the translation messed up due to page breaks, section breaks, hyphenations, page numbers, headers, Footnotes, etc. that will invariably result from using a pdf as your source. If you cannot find your text rendered into any of the above mentioned formats, you'll have to toss your PDF into ABBYY FineReader. Here's a torrent to the full, portable version (which conviently supports old fonts in multiple languages)!

If you don't already have it, download & install uTorrent 2.2.1 Build 25302
Then download the following torrent and click on the portable installer.
ABBYY FineReader PDF 15.0.114.4683 (18.08.2022) (2021) PC | RePack & Portable.
That program should help you tremendously in cases where obtaining a PDF or, God forbid, a poorly scanned PDF of a foreign, handmade book, set with old font type, is the only option available.

I've been doing this kind of thing for nearly a decade now. Let me know if you need any further assistance. I'm always willing to pass down all of the various tricks and tips I picked up along the way as I never had a mentor who could've prevented me from wasting countless hours of inefficiency and wasted time when I first started doing this lol

I can help you do things like remove various DRMs, access public & invite-only ebook websites, utilize handy macros for use in Microsoft Word, etc.

Good luck with your project!
What about having to sign in with details?
What about having to sign in with details?
I tried it out. The google part accepted the document ok 4MB but superimposed the English tranbslation onto the original text in the pdf then the Adobe denied me access to "the cache". Presumably it couldn't separate the texts out ...
So I failed in my trial
What about having to sign in with details?

I tried it out. The google part accepted the document ok 4MB but superimposed the English tranbslation onto the original text in the pdf then the Adobe denied me access to "the cache". Presumably it couldn't separate the texts out ...
So I failed in my trial
Also the Adobe was a very large file + 400MB
 
Last edited:
What about having to sign in with details?

I tried it out. The google part accepted the document ok 4MB but superimposed the English tranbslation onto the original text in the pdf then the Adobe denied me access to "the cache". Presumably it couldn't separate the texts out ...
So I failed in my trial

Also the Adobe was a very large file + 400MB
Update:
I've had some problems with "something"interfering in a major way- with my browser settings.
Just a thought ... surely it's a co-incidence
 
I don't recommend relying on google translate for accuracy. I've seen it get many a thing very wrong, and it often makes a lot of biased assumptions based on what it thinks the context of the submitted text is about.

I do still use google translate sometimes, but I make sure to take a good look at what I'm getting back. I've had better results with DeepL Translate: The world's most accurate translator , but even that isn't perfect, and has far fewer languages to translate from (no Latin, for example).

edit: ew, I definitely wouldn't have titled my hyperlink with their marketing jargon... forum automatically took that from the page title after I posted.
 
What about having to sign in with details?

I tried it out. The google part accepted the document ok 4MB but superimposed the English tranbslation onto the original text in the pdf then the Adobe denied me access to "the cache". Presumably it couldn't separate the texts out ...
So I failed in my trial

Also the Adobe was a very large file + 400MB
Well, as I previously stated:
Try to find source material that is either in epub, docx, mobi, htm(l), azw3, or fb2. That way you can just convert it to docx, toss it into Google translate, and fix the few remaining translation errors, thereby retaining the original chapters, Footnotes, etc, all while not having the translation messed up due to page breaks, section breaks, hyphenations, page numbers, headers, Footnotes, etc. that will invariably result from using a pdf as your source.
I don't recommend relying on google translate for accuracy. I've seen it get many a thing very wrong, and it often makes a lot of biased assumptions based on what it thinks the context of the submitted text is about.

I do still use google translate sometimes, but I make sure to take a good look at what I'm getting back. I've had better results with DeepL Translate: The world's most accurate translator , but even that isn't perfect, and has far fewer languages to translate from (no Latin, for example).

edit: ew, I definitely wouldn't have titled my hyperlink with their marketing jargon... forum automatically took that from the page title after I posted.
Crap! I had so much text typed out on my crappy phone but I stayed on another tab for too long and poof; Page refreshed! Anyways, long-story short, and as much as I hate to say it, for modern language translations of an entire eBook, Google wins, (yes, even despite it's rare idiosyncrasies, Google still has far less of these occurrences than DeepL, and this is coming from someone who [formerly] paid a yearly subscription to DeepL!).

Truth be told, if you have the patience for it... ChatGPT4 all the way! But you've got to be really dedicated as it can only output approximately 5,000-7,500 words for each prompt.
For example, I'm working on both a prose and verse version of Konrads von Würzburg's "Trojanischer Krieg," one section at a time.
Google et al. can only translate modern languages, they cannot translate dead languages.
Here's just a random promt I gave it from The Rape of Helena (which as a side note was almost impossible to find in plain text!)

Me:
Translate the following text into rhymed English verse, ala Thomas Carlyle, trying your best to have your output's word count come as close to the input word count so as not to diverge too far from the original text's authenticity (within reason):
"Cassander dise rede treip.
ân alle fröude si beleip
und tet sich hôchgemüetes abe.
ir jâmer unde ir ungehabe
die wurden alsô bitter,
daz vrouwen unde ritter
mit ir beswæret wâren,
dâ von Prîant die clâren
hin ab dem wege füeren hiez.
in ein gaden er si stiez
und bat si drîn besliezen,
wan in begunde erdriezen
ir clegelîche swære.
man wânde, daz si wære
unsinnic worden bî der stunt,
dô wart in allen schiere kunt,
daz si vil schône sich versan,
wan ir rede sich began
bewæren in vil kurzen tagen.
ir lôzen unde ir wîssagen
erfüllet wurden ûf ein ort.
man spurte schiere, daz ir wort
belîben muosten ungelogen.
diu mære wâren dâ geflogen
über al der Kriechen lant,
daz mit gewalteclicher hant
Helêne was gezücket.
dâ von wart dâ verdrücket
fröud unde hôchgemüete."


ChatGPT4:
In wisdom's grip, Cassander saw
A future where joy fled in awe.
He stepped aside, with spirit high,
As grief and chaos filled the sky.
So bitter was their common fate,
That knights and ladies, once so great,
Were burdened with a weight immense,
Prompting Priam's wise defense.
From the path, he steered them clear,
Into a garden, drawing near.
There he bade them to remain enclosed,
For their lamenting had him opposed.
So heavy was their shared despair,
Some thought madness sparked the air.
Yet soon, to all, it was made known,
Their minds stood firm, like age-old stone.
In days but few, their foresight shown,
Their prophecies, in truth, had grown.
Proven right, their words took flight,
Echoing through the Grecian night.
Tales of Helen, forcefully taken,
Spread o'er lands, by grief awoken,
Dulling joy and pride once spoken.
 
Last edited:
Well, as I previously stated:
Try to find source material that is either in epub, docx, mobi, htm(l), azw3, or fb2. That way you can just convert it to docx, toss it into Google translate, and fix the few remaining translation errors, thereby retaining the original chapters, Footnotes, etc, all while not having the translation messed up due to page breaks, section breaks, hyphenations, page numbers, headers, Footnotes, etc. that will invariably result from using a pdf as your source.

Crap! I had so much text typed out on my crappy phone but I stayed on another tab for too long and poof; Page refreshed! Anyways, long-story short, and as much as I hate to say it, for modern language translations of an entire eBook, Google wins, (yes, even despite it's rare idiosyncrasies, Google still has far less of these occurrences than DeepL, and this is coming from someone who [formerly] paid a yearly subscription to DeepL!).

Truth be told, if you have the patience for it... ChatGPT4 all the way! But you've got to be really dedicated as it can only output approximately 5,000-7,500 words for each prompt.
For example, I'm working on both a prose and verse version of Konrads von Würzburg's "Trojanischer Krieg," one section at a time.
Google et al. can only translate modern languages, they cannot translate dead languages.
Here's just a random promt I gave it from The Rape of Helena (which as a side note was almost impossible to find in plain text!)

Me:
Translate the following text into rhymed English verse, ala Thomas Carlyle, trying your best to have your output's word count come as close to the input word count so as not to diverge too far from the original text's authenticity (within reason):
"Cassander dise rede treip.
ân alle fröude si beleip
und tet sich hôchgemüetes abe.
ir jâmer unde ir ungehabe
die wurden alsô bitter,
daz vrouwen unde ritter
mit ir beswæret wâren,
dâ von Prîant die clâren
hin ab dem wege füeren hiez.
in ein gaden er si stiez
und bat si drîn besliezen,
wan in begunde erdriezen
ir clegelîche swære.
man wânde, daz si wære
unsinnic worden bî der stunt,
dô wart in allen schiere kunt,
daz si vil schône sich versan,
wan ir rede sich began
bewæren in vil kurzen tagen.
ir lôzen unde ir wîssagen
erfüllet wurden ûf ein ort.
man spurte schiere, daz ir wort
belîben muosten ungelogen.
diu mære wâren dâ geflogen
über al der Kriechen lant,
daz mit gewalteclicher hant
Helêne was gezücket.
dâ von wart dâ verdrücket
fröud unde hôchgemüete."


ChatGPT4:
In wisdom's grip, Cassander saw
A future where joy fled in awe.
He stepped aside, with spirit high,
As grief and chaos filled the sky.
So bitter was their common fate,
That knights and ladies, once so great,
Were burdened with a weight immense,
Prompting Priam's wise defense.
From the path, he steered them clear,
Into a garden, drawing near.
There he bade them to remain enclosed,
For their lamenting had him opposed.
So heavy was their shared despair,
Some thought madness sparked the air.
Yet soon, to all, it was made known,
Their minds stood firm, like age-old stone.
In days but few, their foresight shown,
Their prophecies, in truth, had grown.
Proven right, their words took flight,
Echoing through the Grecian night.
Tales of Helen, forcefully taken,
Spread o'er lands, by grief awoken,
Dulling joy and pride once spoken.

This very interesting. Do you have more?

The section above seems to signal a mythic-poetic confirmation of KD's hypothesis that Helen is a light technology.
 
Tips
Tips
Please respect our Posting Rules.
Back
Top