Problem: Creating Translation Memory with OmegaT Tópico cartaz: Andy S
|
|---|
Andy S
Reino Unido
Local time: 04:18
alemão para inglês
I am doing a De>En project today with no TM available. The project I alluded to yesterday came with a TM supplied by the client.
However, there are related ST and TT files stored on disk from previous projects. I take it that OmegaT will be no use to me with no TM available - copying TT files into the TM directory os pointless, right? So I have to do this job, save the TM and then work with it with future similar jobs?
[Subject edited by staff or moderator 2008-11-06 11:16]... See more I am doing a De>En project today with no TM available. The project I alluded to yesterday came with a TM supplied by the client.
However, there are related ST and TT files stored on disk from previous projects. I take it that OmegaT will be no use to me with no TM available - copying TT files into the TM directory os pointless, right? So I have to do this job, save the TM and then work with it with future similar jobs?
[Subject edited by staff or moderator 2008-11-06 11:16]
BTW have now downloaded Bitext2tmx, now I have to work out how to open the archive!
[Edited at 2008-11-06 18:22] ▲ Collapse
| | | | Samuel Murray 
Holanda
Local time: 05:18
Membro (2006)
inglês para africâner
+ ...
| Three options | Nov 6, 2008 |
Andy S wrote:
However, there are related ST and TT files stored on disk from previous projects. I take it that OmegaT will be no use to me with no TM available - copying TT files into the TM directory os pointless, right? So I have to do this job, save the TM and then work with it with future similar jobs?
You can do three things, none of which are very satisfactory.
1. Align your old translations using one of the aligners listed on OmT's web site, or any other aligner (eg PlusTools). This takes long but is the most useful.
2. Put all these files of yours in a single folder, and use the reference search feature in OmegaT. This takes only a minute, but you need to do the searching manually.
3. Combined with #2: Create a dummy project with your SL files, then set your OmT settings to copy source to target and to add such segments to the TM, then run through all the segments, so that you at least have the SL text in one large TM. Put that TM in your TM folder. If you get matches from it, you'll know to do #2 above for that particular phrase.
| | | | Marc P (X)
Local time: 05:18
alemão para inglês
+ ...
| Creating Translation Memory with OmegaT | Nov 6, 2008 |
A couple of points to add to Samuel's good advice:
Manual alignment can be quite time-consuming and is generally only worthwhile for texts that are very similar (e.g. the TM relates to a previous version of the document you are now working on).
If you do decide to align manually, opt for paragraph-level rather than sentence-level segmentation in the aligner. This provides both more accurate alignment, and fewer segments overall to align.
As Samuel says, yo... See more A couple of points to add to Samuel's good advice:
Manual alignment can be quite time-consuming and is generally only worthwhile for texts that are very similar (e.g. the TM relates to a previous version of the document you are now working on).
If you do decide to align manually, opt for paragraph-level rather than sentence-level segmentation in the aligner. This provides both more accurate alignment, and fewer segments overall to align.
As Samuel says, you can search across legacy documents. Consider using a desktop indexing tool such as dtSearch (Windows) or Beagle (Linux) if you don't already have one - this is likely to be faster than OmegaT's integral file search facility, unless you strictly limit the scope of search.
Marc ▲ Collapse
| | | | Vito Smolej
Alemanha
Local time: 05:18
Membro (2004)
inglês para esloveno
+ ...
SITE LOCALIZER | Clearing questions | Nov 6, 2008 |
.... The project I alluded to yesterday came with a TM supplied by the client.... there are related ST and TT files stored on disk from previous projects. I take it that OmegaT will be no use to me with no TM available...
Is there any bilingual material available for the project? If there's a good chance for it, then the next question is "in what form" - for instance in what form is the TM "supplied by the client"? If it is Trados-based, I can help. If it is corresponding source and target files, Samuel gave you a good hint.
Regards
Vito
PS: the questions are generic, i.e. they have to be asked whatever the CAT tool used.
| | |
|
|
|
Andy S
Reino Unido
Local time: 04:18
alemão para inglês
CRIADOR(A) DO TÓPICO | No TM for this one | Nov 6, 2008 |
Thanks for the help, gentlemen...there is no TM supplied by the client for this latest project -but a couple of of bilingual files from legacy translations of mine, pre-OmegaT.
I will try out the methods Samuel suggested and see how I get on.
With thanks
Andrew
| | | | Andy S
Reino Unido
Local time: 04:18
alemão para inglês
CRIADOR(A) DO TÓPICO | Open archive? | Nov 8, 2008 |
What programme do I need to open the bitext2.tmx archive please?
| | | | Marc P (X)
Local time: 05:18
alemão para inglês
+ ...
| Open archive? | Nov 8, 2008 |
I am not sure what you mean by "open the archive".
Bitext2tmx is a program that takes two files (a source text and its translation) and aligns their content. The files must be in plain-text format, so you must first convert your files to plain text if necessary. Bitext2tmx segments the two files and presents their content, segment by segment and side-by-side in a table in which you can correct any misalignment. After correcting the misalignment, you save the result, which Bitext2tmx... See more I am not sure what you mean by "open the archive".
Bitext2tmx is a program that takes two files (a source text and its translation) and aligns their content. The files must be in plain-text format, so you must first convert your files to plain text if necessary. Bitext2tmx segments the two files and presents their content, segment by segment and side-by-side in a table in which you can correct any misalignment. After correcting the misalignment, you save the result, which Bitext2tmx then outputs in the form of a .tmx file.
Marc ▲ Collapse
| | | | esperantisto
Local time: 07:18
Membro (2006)
inglês para russo
+ ...
SITE LOCALIZER | Don't open it, | Nov 9, 2008 |
but run as a regular Java program.
| | |
|
|
|
Didier Briel
França
Local time: 05:18
inglês para francês
+ ...
| The one matching the extension | Nov 9, 2008 |
Andrew Stucken wrote:
What programme do I need to open the bitext2.tmx archive please?
On Sourceforge, bitext2tmx comes with a number of extensions. You need a compatible archive manager.
If you downloaded bitext2tmx-1.0M0-080229.7z, for instance, you can use 7-Zip
Didier
| | | | Andy S
Reino Unido
Local time: 04:18
alemão para inglês
CRIADOR(A) DO TÓPICO | Não há um moderador designado especificamente para este fórum. Para reportar violações às regras do site ou para obter ajuda, favor contatar a equipe do site » Problem: Creating Translation Memory with OmegaT | 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 » |
| | Pastey | Your smart companion app
Pastey is an innovative desktop application that bridges the gap between human expertise and artificial intelligence. With intuitive keyboard shortcuts, Pastey transforms your source text into AI-powered draft translations.
Find out more » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
