Question about CafeTran > TMX and TMX > CafeTran
Thread poster: MikeTrans
MikeTrans
Germany
Local time: 08:58
Member (2005)
Italian to German
+ ...
Jun 19, 2012

Hi,

1)
CafeTran > TMX:

I have just finished a project with a single document and I want to polish my TMX segment memory, seeing that it contains some doubles, segments not in the natural order of the translated document etc..

The simplest way is just to delete from my memory the first to the last added segments (with a third-party program), and then to send my entire project segments to the memory again (or into a new memory).

Can I do that? How?

2)
TMX > CafeTran:

For example, after doing some QA and reviewing outside of CafeTran, I have an updated TMX file and want to get it into my CafeTran project again.

So, Is this the correct way? :

Edit > Task > Remove target segments;
pretranslation using the updated memory with Translation > Pretranslate Project segments


Thanks for feedback,
Mike


Direct link Reply with quote
 
MikeTrans
Germany
Local time: 08:58
Member (2005)
Italian to German
+ ...
TOPIC STARTER
1) is: Save Project as... Jun 19, 2012

Hi,

Difficult to find this one
"Save Project as..." can save a project in HTML, TMX , TXT, not just save to another name (as I thought).

As for pretranslation:
Translation > Translate a document,
Translation > Translate Project segments

both show no activity. Any special things to carry out first?

Mike


[Edited at 2012-06-19 18:51 GMT]


Direct link Reply with quote
 

Igor Kmitowski  Identity Verified
Poland
Local time: 08:58
English to Polish
+ ...
Pretranslation Jun 20, 2012

The Pretranslation function in CafeTran is a background process which should be used when working with huge memories. After it kicks off, the program translates the segments in the background, and you can just go to the next segment to check the result. You don't have to wait until Pretranslation is over and the response/recognition should be instant without any delay when you go to the next segment.

If you want to fill the target segments with the TM matches automatically, then you need to wait until Pretranslation is over (see the progress bar) and select any of the Jump over options in the Edit menu.

Igor

MikeTrans wrote:

pretranslation using the updated memory with Translation > Pretranslate Project segments


Thanks for feedback,
Mike


[Edited at 2012-06-20 08:45 GMT]


Direct link Reply with quote
 

Meta Arkadia
Local time: 14:58
English to Indonesian
+ ...
Import it in your TMX, I'd say Jun 20, 2012

Well, if you do the proofing as suggested by the other Hans in his KB (http://cafetran.wordpress.com/2012/06/19/how-to-proofread-a-translation-made-in-cafetran), save the proofed version as a tab del file, open it as a glossary in your project, and import it (not saving duplicates) in your TM you can pretranslate your document.

Two problems: any tags will disturb the process, and CT can't pretranslate in one go, only segment for segment. Igor doesn't think this is a priority, I think it is.


Direct link Reply with quote
 

Igor Kmitowski  Identity Verified
Poland
Local time: 08:58
English to Polish
+ ...
Jump over Jun 20, 2012

What you mean by 'pretranslate in one go' is to have the targets segments filled with TM matches. Right? So, if the Pretranslation process is over, make sure that Jump over | Perfect Memory Matches and Autotranslation and Fuzzy matches are ticked. Then, CafeTran will autoinsert fuzzy and perfect matches 'in one go'.

Igor

Meta Arkadia wrote:

Well, if you do the proofing as suggested by the other Hans in his KB (http://cafetran.wordpress.com/2012/06/19/how-to-proofread-a-translation-made-in-cafetran), save the proofed version as a tab del file, open it as a glossary in your project, and import it (not saving duplicates) in your TM you can pretranslate your document.

Two problems: any tags will disturb the process, and CT can't pretranslate in one go, only segment for segment. Igor doesn't think this is a priority, I think it is.


Direct link Reply with quote
 
MikeTrans
Germany
Local time: 08:58
Member (2005)
Italian to German
+ ...
TOPIC STARTER
Igor, Meta, ... Jun 20, 2012

... thank you very much for your tips.

Igor,

I find this pretranslation system very intelligent as it saves a lot of computer ressources.
By trying out, I found it very practical to load big TMs using only the Fuzzy mode, or load them in full Fuzzy + sub-segment mode by loading only a filtered part of them. Works very well as now I have x64 Java installed. Also I will consider to upgrade my Laptop to 8 GB Ram. This should close for a long time any performance questions.

I was able to pre-translate a document by just opening the doc in a new project and load a segment memory. I chose "Translate project segments" and then I press the Start button in the target area.

CT immediately goes through the segments and fills all target segments.
I imagine, if I set the Fuzzyness from 33 to 70, CT would also fill the segments with Fuzzy matches appropriately.

Meta,

I'm just trying out now a docx document which is 'tagged', that is formated, somewhat above average.
I will run the "Progressive segmentation" to see and learn about tags in CT, but for the future, I can imagine that having all sorts of documents and memories exported to TMX, there is an advantage if you do some migration operations (send your project to another CafeTran user, send a TMX to a colleague which uses a different CAT tool):

With TMX, you are almost *guaranteed* to maintain segmentation from one CAT to another.
As for the tags, this is more tricky, they should all be written in the TMX file and *stay* there, but I will consider writing a macro or a longer script in UltraEdit (a text editor) in order to convert tags from Trados Studio, TTX > CafeTran and vice-versa. If I succeed, I can finally forget about the nuisance of handling TTX files. Converting tags from CT > DVX2 / MemoQ > DVX2 is far more simple, as they all use progressive trailing numbers.

Optimally, I can have *all* my ressources in TMX for working seriously without wasting time: With CT you don't even need to have glossaries, term memories will do the job and you can set the stemming and fuzziness.)

Mike


Direct link Reply with quote
 


To report site rules violations or get help, contact a site moderator:

Moderator(s) of this forum
Natalie[Call to this topic]

You can also contact site staff by submitting a support request »

Question about CafeTran > TMX and TMX > CafeTran

Advanced search






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 »
LSP.expert
You’re a freelance translator? LSP.expert helps you manage your daily translation jobs. It’s easy, fast and secure.

How about you start tracking translation jobs and sending invoices in minutes? You can also manage your clients and generate reports about your business activities. So you always keep a clear view on your planning, AND you get a free 30 day trial period!

More info »



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