Pages in topic:   [1 2] >
Importing a translation memory into OmegaT
Thread poster: Michael Mestre

Michael Mestre
France
Local time: 05:49
English to French
+ ...
Sep 29, 2009

Hi all,

I have two questions about Translation Memories in Omega T that may sound trivial, but I haven't been able to find a solution by myself :

1) How can I load a standard TMX file into my OmegaT project ?
For example, one that has been created by another program (stingray, ...)

2) How can I merge a TM file (the native OmegaT TMX format, or a standard TMX file) into my current TM for a project that I have already started working on ?
For example, I start working on a project (Project2) which is about telecommunications. Then after a while I realize that there is a small part with a lot of terms that I have already translated in the past, in a project (Project1) about routers.
I wish to incorporate Project1's TM into my current TM for Project2 in order to see the fuzzy matches.

I am sure that there is a simple answer to the first question, and there is probably a way to handle the second.
Any ideas ?


 

Laurent KRAULAND  Identity Verified
France
Local time: 05:49
French to German
+ ...
OmegaT's help Sep 29, 2009

Hi Michael,
this is what the OmegaT handbook says:
Importing files
OmegaT supports importing TMX versions 1.1-1.4b (both level 1 and level 2). This enables the translation memories produced by other tools to be read by OmegaT. However, OmegaT does not fully support imported level 2 TMX files (that store not only the translation, but also the formatting), so the quality of fuzzy matches will be a bit lower when importing level 2 TMX files.

Imported files handling
OmegaT follows very strict procedures when handling translation memory files (TMX files). If any mistake is found in such a file, OmegaT will simply not load it.
Some tools are known to produce invalid TMX files in certain conditions. If you want to use such files as reference translations in OmegaT, they will need to be fixed if OmegaT refuses them. Fixes are trivial operations and OmegaT will give you indications in the related error message. You can ask the user group for advice if you have problems.


I am not currently working on projects with OmegaT, but assume you can import a TMX file in the right format (see above) by using the "Project" tab of OmegaT.

As per merging TM's, I think there could be a rather obvious workaround (or even procedure) to be followed to do this.

[Edited at 2009-09-29 16:24 GMT]


 

Robert Tucker
United Kingdom
Local time: 04:49
German to English
+ ...
Further from OmegaT Help Sep 29, 2009

Should you wish to reuse a translation memory from a previous project (for example because the new project is similar to the previous project, or uses terminology which might have been used before), you can use these translation memories as "input translation memories", i.e. for import into your new project. In this case, place the translation memories you wish to use in the \tm folder of your new project.

Note "memories" plural – so you don't need to merge TM's to use them with OmegaT, you can place as many as you like in the \tm folder.


 

Susan Welsh  Identity Verified
United States
Local time: 23:49
Member (2008)
Russian to English
+ ...
tmx Sep 29, 2009

Just stick the tmx file into your OmegaT project folder /tm
I don't know anything about merging.


 

Michael Mestre
France
Local time: 05:49
English to French
+ ...
TOPIC STARTER
Thanks Laurent ! Sep 29, 2009

Laurent, thank you for your suggestions. From the Project tab of OmegaT, and by looking at the directories, I guess that I can dump all the TM's that I want in the TM/ folder; is this correct ?

Will multiple files be loaded by doing this ? If this is the case, then it more or less answers my question about merging TMs..


 

Laurent KRAULAND  Identity Verified
France
Local time: 05:49
French to German
+ ...
I am not sure.... Sep 29, 2009

about replying Yes or No to your question, Michael... But at in other projects, I assume that OmegaT will only accept one single file to serve as a TM.
Anyway, I would be more than glad (on a reasonable basis, of course) to convert and merge TM's so they can be compatible with OmegaT for colleagues asking for it. Contact data are to be found on http://www.krauland.tel.
BTW where are Samuel and Vitoicon_eek.gif?

[Edited at 2009-09-29 17:27 GMT]


 

Samuel Murray  Identity Verified
Netherlands
Local time: 05:49
Member (2006)
English to Afrikaans
+ ...
Another answer Sep 29, 2009

Michael Mestre wrote:
1) How can I load a standard TMX file into my OmegaT project ?
For example, one that has been created by another program (stingray, ...)


Two possible ways:

1. You can save it in the /tm/ folder of the project. Then OmegaT will serve fuzzy matches from it, but 100% matches will not be inserted into your translation automatically. You can have an unlimited number of TMX files in the /tm/ folder, and OmegaT will consult them all.

2. You can rename it to "project_save.tmx" and save it in the /omegat/ folder. Then OmegaT will serve fuzzy matches from it, and 100% matches will be inserted into your translation automatically. You can only have one TM called "project_save.tmx" in your project -- it is the main TM, from which your translated files will be created.

2) How can I merge a TM file (the native OmegaT TMX format, or a standard TMX file) into my current TM for a project that I have already started working on ?


See answer above. But if you truly want to merge TMs into a single TM, you'd have to use a separate TMX merging utility (or merge them manually using copy/paste in a good text editor).

I wish to incorporate Project1's TM into my current TM for Project2 in order to see the fuzzy matches.


Yes, if you only want to see fuzzy matches, and you don't mind having to press a keyboard shortcut to accept 100% matches, then option #1 above is for you.


 

Samuel Murray  Identity Verified
Netherlands
Local time: 05:49
Member (2006)
English to Afrikaans
+ ...
Another answer II Sep 29, 2009

Michael Mestre wrote:
From the Project tab of OmegaT, and by looking at the directories, I guess that I can dump all the TM's that I want in the TM/ folder; is this correct ?


Yes.

Will multiple files be loaded by doing this ? If this is the case, then it more or less answers my question about merging TMs..


Yes. All TMX files in the /tm/ folder are read, and OmegaT will serve the top 5 fuzzy matches from among all of them.


 

Michael Mestre
France
Local time: 05:49
English to French
+ ...
TOPIC STARTER
Thanks to everyone Sep 29, 2009

Thanks a lot again to everyone (and to Samuel) for your very valuable help. You have answered all my questions !

 

Fabian Luttman  Identity Verified
Canada
Local time: 23:49
Member (2003)
English to Spanish
+ ...
Main TM file Jan 18, 2014

Hello Samuel,

I am new with OmegaT, and I am reading all I can find in order to learn its tricks.

I have a question regarding one of your suggestions:

2. You can rename it to "project_save.tmx" and save it in the /omegat/ folder. Then OmegaT will serve fuzzy matches from it, and 100% matches will be inserted into your translation automatically. You can only have one TM called "project_save.tmx" in your project -- it is the main TM, from which your translated files will be created.


This means that there will be ONLY ONE main TM?
What happens if you usually translate in more than one language combination?
Is there any way of having at least one main TM per language pair?

Thanks,

Fabian



[Edited at 2014-01-18 17:05 GMT]

[Edited at 2014-01-18 19:46 GMT]


 

Samuel Murray  Identity Verified
Netherlands
Local time: 05:49
Member (2006)
English to Afrikaans
+ ...
OmegaT doesn't know... Jan 18, 2014

Fabian Luttman wrote:
This means that there will be ONLY ONE main TM?
What happens if you usually translate in more than one language combination?
Is there any way of having at least one main TM per language pair?


1. OmegaT doesn't know into which language every individual source document you want to translate into. If the project's language codes are e.g. EN>ES then OmegaT assumes that your source documents are all English (even if they're not) and assumes that you want to translate them into Spanish (even if the actual text that you type into OmegaT is not Spanish). If you have two source documents at the same time and you want one translated into Spanish and one into e.g. Russian, then you have to create two separate projects for it in OmegaT.

2. I'm not sure if you understand what was meant by "main TM" in the original discussion. The thing that we called the "main TM" in this forum thread (project_save.tmx) is just a temporary TM that is created while you do the translation. You can do all sorts of tricks with it, of course, but for normal working conditions you're not supposed to do anything with the "main TM" at all. As soon as you've finished with your translation and told OmegaT to create the target files, OmegaT exports all used segments to a separate TM (in three different formats, for your convenience), and after you've moved those TMs to whatever location on your computer on your hard drive where you store your old TMs, you can delete the project folders (or zip it up for backup purposes) (including the project's "main TM".


 

Didier Briel  Identity Verified
France
Local time: 05:49
Member (2007)
English to French
+ ...
There's also the auto folder Jan 18, 2014

Since people are referring to this old thread.

Samuel Murray wrote:

Michael Mestre wrote:
1) How can I load a standard TMX file into my OmegaT project ?
For example, one that has been created by another program (stingray, ...)


Two possible ways:

1. You can save it in the /tm/ folder of the project. Then OmegaT will serve fuzzy matches from it, but 100% matches will not be inserted into your translation automatically. You can have an unlimited number of TMX files in the /tm/ folder, and OmegaT will consult them all.

2. You can rename it to "project_save.tmx" and save it in the /omegat/ folder. Then OmegaT will serve fuzzy matches from it, and 100% matches will be inserted into your translation automatically. You can only have one TM called "project_save.tmx" in your project -- it is the main TM, from which your translated files will be created.

Inside the /tm/ folder, you can have an /auto/ folder. OmegaT will insert automatically exact matches coming from the /tm/auto/ folder.


 

Fabian Luttman  Identity Verified
Canada
Local time: 23:49
Member (2003)
English to Spanish
+ ...
Still about TMs... Jan 18, 2014

Thank you for the quick responses!
I thought that this old thread might be dead already, but I was wrong!!

Here is my specific problem:
I am an advanced user of Trados Studio, and I am trying to make my way out of it and into OmegaT.
Why?
Well, there is not a single reason...
I am a reasonably proficient Linux user (Ubuntu) and I really hate to have to run a virtual Windows machine ONLY to be able to use Trados.
I have been using Trados since its 2007 version, and it gets bulkier and slower with every new version coming out.
I won't even comment on Trados price strategy...
I could go on, but it's not the point of my question.

So far, I am just learning the basics of omegaT, and I think it could be a good replacement for Trados.
One thing I got used to while using Trados is the capability of choosing the language pair (hence the TM) when you start translating a file or project.
Is there any way to do something similar in omegaT?
I have 4 or 5 main TMs that I need to use almost every day, for a couple of different language combinations and for different clients. I have them all in the same folder, identifiable by their names.
What would be the most efficient workflow in omegaT in order to select one of those main TMs at the beginning of a project?
I don't like the idea of having to copy, move or rename TM files before starting every single project...
I prefer Trados approach to this: easily selecting the TM you want to use for a particular project (assuming you already have one), working on that project using fuzzy and 100% matches, and updating that TM at the same time so it will include your latest work to be used next time around.


Thank you for your help!

Fabian

[Edited at 2014-01-18 19:48 GMT]

[Edited at 2014-01-18 20:25 GMT]

[Edited at 2014-01-19 15:00 GMT]


 

esperantisto  Identity Verified
Local time: 06:49
Member (2006)
English to Russian
+ ...
Well… Jan 19, 2014

Fabian Luttman wrote:

I prefer Trados approach to this: easily selecting the TM you want to use for a particular project (assuming you already have one), working on that project using fuzzy and 100% matches, and updating that TM at the same time so it will include your latest work to be used next time around.


I don’t deem such a strategy efficient (the bigger is TM, the greater effort is required for its maintenance), but if you do, simply create several dedicated projects, set your big TMs as respective project main TMs, and proceed by adding new files to to the projects and removing old files as necessary.


 

Fabian Luttman  Identity Verified
Canada
Local time: 23:49
Member (2003)
English to Spanish
+ ...
Regarding TMs Jan 19, 2014

esperantisto wrote:

I don’t deem such a strategy efficient (the bigger is TM, the greater effort is required for its maintenance), but if you do, simply create several dedicated projects, set your big TMs as respective project main TMs, and proceed by adding new files to to the projects and removing old files as necessary.


Hello esperantisto,

If I understand correctly, this will imply making a copy of my main TM per every project I want to work on. Not exactly what I want to do...
Regarding the size of the TM, if we agree that having a TM is a good idea in the first place, I can't see why having a big TM will become a bad idea. The bigger the TM, the better. Of course it will require a bigger effort to maintain it, but that is usually the case with everything.
I got used to the way Trados manages it. I am not saying it is the best way, all I am saying is that I like it, I got used to it, and I don't think it is a terrible idea from the logistics point of view...
You have all your TMs (for different languages, different subjects, different clients, etc.) in one folder called TMs, you choose which one you want to use when you start a new project, you use it along the project and it will automatically grow as you work with it so it will be ready and up to date when you start your next project. I can't see anything seriously wrong with this workflow.
Regarding maintenance, I like the opportunity that Trados gives you to edit/delete a unit, if necessary, during the translation process. That helps in keeping thing under control without having to dedicate specific time for its maintenance. But again, it's all about preferences and what you have got used to... With time, new and better things will show up, but humans have resistance to changes in general. When we get used to do something in a specific way we get convinced that it is the best way, and in many, many cases it proved not to be the case.icon_wink.gif

Fabian


 
Pages in topic:   [1 2] >


There is no moderator assigned specifically to this forum.
To report site rules violations or get help, please contact site staff »


Importing a translation memory into OmegaT

Advanced search






BaccS – Business Accounting Software
Modern desktop project management for freelance translators

BaccS makes it easy for translators to manage their projects, schedule tasks, create invoices, and view highly customizable reports. User-friendly, ProZ.com integration, community-driven development – a few reasons BaccS is trusted by translators!

More info »
Déjà Vu X3
Try it, Love it

Find out why Déjà Vu is today the most flexible, customizable and user-friendly tool on the market. See the brand new features in action: *Completely redesigned user interface *Live Preview *Inline spell checking *Inline

More info »



Forums
  • All of ProZ.com
  • Term search
  • Jobs
  • Forums
  • Multiple search