Software for non-technical translation
Thread poster: menisk
I'm not a professional and I need your advice. I'm trying to translate a certain biography as a hobby project and it's a bit of a pain software-wise. Using a generic text editor while peeking in an open paper book couldn't be the best way, but I fail to understand if there are better tools for manual translation. I don't need translation memory (it is of no use), just a tool that would
1) Align the original (I have it scanned and ocr'ed) and translation. I'm thinking about a two-pane editor or something with tables maybe.
2) Indicate the parts that are not translated yet.
Seems very intuitive and basic, but somehow I can only find either huge CATs aimed at technical texts or software resource file translation utilities (maybe I should just try one of those?). Perhaps I don't understand the workflow.
What do people use for such content?
[Edited at 2011-10-15 08:15 GMT]
| Use a CAT tool || Oct 15, 2011 |
I would strongly recommend using a CAT tool for this purpose. Many people think that such tools are only relevant for technical translation using translation memory. Not so. One of the major advantages is that such tools is that the display the source and target texts side by side (or one above the other). This I find is far superior to the traditional overwriting method. Very much like the two pane solution you are looking for.
As you translate, it is easy to tell how much You have done, and how much remains to be done.
Used iike that, CAT tools are pretty intuitive and basic. They meet your requirements, though perhaps not in quite the fashion you expected.
As for which CAT tool, a simple cheap one with an intuitive display is all you need.
| | James McVay
Local time: 04:29
Russian to English
That is exactly why I use a CAT tool. If I don't, I tend to skip whole sentences.
You might like to try Wordfast Anywhere or Google Translator Toolkit. Nothing to download -- they're online tools and simple to use. If you're working with a hard copy document, you'll need to OCR it, of course. Here are the URLs:
Google Translator Tookit:
Local time: 10:29
French to English
| Use of CAT tools for literary translations || Oct 15, 2011 |
This topic has been discussed at great length in several other threads, but I just felt it was important here to add my voice as a non-user of CAT tools.
The point raised by the immediately-preceding contributors is the very reason why I don't like using CAT tools for this kind of work — the fact that the CAT tool 'segments' the text (= breaks it down into working chunks') on the basis of (usually) relatively simple rules means that, by an large, you are presented with the text to be translated sentence by sentence (on the basis of the presence of a full-stop/period followed by a space); sub-sentence segmentation would not generall be a very workable option for this kind of work, I don't think.
Now I often find that for the sake of a more natural style, or the flow of ideas, or a number of other reasons, in more 'literary' style translations I quite often want to re-arrange the sentence boundaries, sometimes combining 2 source-text sentences into one target text one, or sometimes, the reverse, splitting a long sentence into 2 or more; or worse still, moving the sentenc bunadries between a whole string of sentences.
Now I can't speak for other CAT tools, but using Wordfast Classic, the process for undoing segmentation and then forcing it to segment the way I want it to is time-consuming, unergonomic, and tends to break my train of thought.
I don't see why you can't simply work by overtyping in a copy of your OCR'ed source file, and then if necessary, for whatever reason, align you source and target texts once you finish (at which point, you would have to deal manually with the hopefully-not-too-numerous re-segmented parts); this is by and large the way I work, and I know that the small amount of work involved in over-typing then deleting is still infinitely faster than arguing the toss with a CAT programme.
We've already established that the TM feature is probably going to be of little use in this type of work, given the lack of repeats; I can't even think that the terminology glossary feature would be much help to you here (though I personally do find that generally one of the most useful features) — if you have a few words that are likely to regularly recur and need translating, such as Londres > London, then it's not very time-consuming to do a search-and-replace-all on such terms as you go along and discover them.
Otherwise, I sometimes simply convert the source text into a table, which will probably separate it at paragraph breaks, thus giving you one paragraph per cell, and then create a second column to the right where I type my translation (tip: I usually copy-and-paste the left-hand column to the right, to give me something to start overtyping on, as all the proper names, for example, will be there ready for me!)
I've tried various different ways of working, but these are the ways that I find most efficient for me.
By the way, I am a fast but sloppy 2-finger typist, and I use keyboard shortcuts wherever possible, setting up custom ones to do certain things I need most often. Hence typing then deleting ahead is really quite fast, and I mix re-typing with leaving word in, some of which can even stay in the source language (in my case FR) where they are near enough in spelling for the spell checker to correct. So, for example, if I have the word 'qualité', I leave it, because at the spell-checking stage, the Word spell-checker will offer me 'quality' and then enable me to 'replace all'; very quick, once you get used to using it and know which words work and which don't! I've found that this alone saves me a great deal of actual fresh typing.
I hope these few comments are of some help.
[Modifié le 2011-10-15 21:28 GMT]
| || |
| To CAT or not to CAT || Oct 16, 2011 |
I use CATs for all my work and I would suppose I am quite advanced in their use, yet I would be hesitant to use them for literary translation. The points Tony raises are quite valid - segmentation into sentences might result in "choppy" translation, so the text might lose its flow (and you might lose your train of thought more easily).
One solution might be to change the segmentation to paragraph level - the paragraph boundaries rarely need to be crossed. However, test out several tools just for the looks - some CATs might have problems displaying such large chunks of text, as they are not often used that way.
| CAT tools revisited || Oct 16, 2011 |
Tony M wrote:
— the fact that the CAT tool 'segments' the text . . means that you are presented with the text to be translated sentence by sentence . . . segmentation would not generally be a very workable option for this kind of work, I don't think.
I entirely agree with you about the need to rearrange sentence boundaries. In Trados at least it is very easy to break up or combine segments, so I do not find that it breaks my train of thought. But I accept that this is a very subjective issue.
I don't see why you can't simply work by overtyping in a copy of your OCR'ed source file
When I started translating 10 years ago, of course I used overtyping. But even then it struck me as a clumsy way of working, prone to making mistakes such as omitting words or sentences. it was a great relief when I discovered that CAT tools enabled me to enter my translation without overwriting the original.
We've already established that the TM feature is probably going to be of little use in this type of work
Agreed. The only real benefit of CAT tools is the display. But that alone justifies their use. Having said that, one occasional benefit of TMs in literary translation is to identify repetitions – precisely in order to avoid them.
By the way, I am a fast but sloppy 2-finger typist
Have you considered using speech recognition? In my experience, SR is ideally suited to literary translation (except poetry), enhancing both quantity and quality.
| || |
To report site rules violations or get help, contact a site moderator:
You can also contact site staff by submitting a support request »
Software for non-technical translation
|Manage your TMs and Terms ... and boost your translation business|
Are you ready for something fresh in the industry? TM-Town is a unique new site for you -- the freelance translator -- to store, manage and share translation memories (TMs) and glossaries...and potentially meet new clients on the basis of your prior work.
More info »
|Anycount & Translation Office 3000|
|Translation Office 3000|
Translation Office 3000 is an advanced accounting tool for freelance translators and small agencies. TO3000 easily and seamlessly integrates with the business life of professional freelance translators.
More info »