Automatic "translation" of numbers
Thread poster: Mikhail Popov

Mikhail Popov
Singapore
Local time: 00:09
Member (2015)
English to Russian
+ ...
Feb 11, 2009

Now I'm trying to evaluate functionality of MemoQ. Everything is really good so far, but I've got some problems.

One of them is automatic "translation" of numbers. If a separate segment contains only digits (and punctuation characters), Wordfast (for example) will skip it and I don't need to translate it at all. MemoQ consider numbers as a segment that should be translated. I understood how to deal with it by autotranslatables, but in this case I have to manually select translation from the box to insert it into the target segment.

But I want MemoQ to skip segments with digits and punctuation characters automatically. I guess I should work with Non-translatables, but couldn't get any results.

Could somebody give me any recommendations?


Direct link Reply with quote
 
István Lengyel
Hungary
Local time: 18:09
English to Hungarian
+ ...
use pre-translation Feb 12, 2009

Hello Michael,

If you don't want to deal with numbers, you can skip them pretty easily.

1. Enter the appropriate auto-translatable rules. (This can help: http://en.wikibooks.org/wiki/CAT-Tools/MemoQ/Tips_and_Tricks#Using_auto-translatables_for_number_format_conversion)
2. Pre-translate the document using fragment assembly. This will include the numbers.
3. If you don't like the fact that the program copied not only numbers-only segments, sort the whole document by segment length using the filtering and sorting toolbar, and remove the translations in the longer segments.
4. If you want, lock the pre-translated segments using Lock special. And if you absolutely want to do so, go to Goto next settings and set the cursor so that it jumps over locked segments.

I hope this helps!

István


Direct link Reply with quote
 

Alexandre Oberlin  Identity Verified
France
Local time: 18:09
English to French
+ ...
Very simple indeed just to keep integers as they are! Nov 15, 2010

Hi all,

István Lengyel wrote:
If you don't want to deal with numbers, you can skip them pretty easily.

Pretty easily really? I have huge lists of bare integers from tables (not even a decimal point) and MemoQ insists in extensively searching the TMs for each of them and expecting me to enter a translation!!!

The boxes are checked in Project home / Settings / Auto translation rules.

I don't want to pretranslate and above all I have no time to lose!

Never seen such a thing...

Any *simple* trick ?

Cheers,

Alexandre


Direct link Reply with quote
 

Mikhail Popov
Singapore
Local time: 00:09
Member (2015)
English to Russian
+ ...
TOPIC STARTER
Simple trick Nov 16, 2010

Simple trick is to mark all these numbers as hidden text, if you are dealing with MS Word.
If you work with Excel table, you can exclude these regions during MemoQ import stage.

Or, you can simply sort all text in memoQ by text length. The shortest segments are usually with numbers only. You can mark the whole range, copy source to target and confirm without updating TM. This could help.


Direct link Reply with quote
 

Alexandre Oberlin  Identity Verified
France
Local time: 18:09
English to French
+ ...
Numbering behaviour seems beyond control Nov 17, 2010

Hi,

Michael Popov wrote:

Simple trick is to mark all these numbers as hidden text, if you are dealing with MS Word.
If you work with Excel table, you can exclude these regions during MemoQ import stage.

Or, you can simply sort all text in memoQ by text length. The shortest segments are usually with numbers only. You can mark the whole range, copy source to target and confirm without updating TM. This could help.


Thank you Michael for this hint. I have to stay inside MemoQ as this is an inx file. Shall I be able to get back to the initial segment order ?

I really get strange things with those numbers. For instance I also have big numbers with commas as separators in the EN source. Then sometimes I get the copy of the source, i.e. the number with commas, which in FR is quite incorrect, or I get anything from a short string or number with some digits in common to an empty cell. Sometimes the automatic match is incorrect (made up from fragments I guess), but I can get a correct number (with spaces, since this is what has become the standard) by explicitly calling the TM match. To put it shortly, it's a complete mess.

I wish software vendors could refrain from introducing features they obviously don't control. Or at least give us an easy way to disable them completely. Would it be too simple just to localize the decimal and thousands separators and otherwise keep the numbers as they are?

That number issue and some other (like the auto translation rules) apart, I should say that I've a good overall feeling working on MemoQ for the first time.

Cheers,

Alexandre


Direct link Reply with quote
 

Mikhail Popov
Singapore
Local time: 00:09
Member (2015)
English to Russian
+ ...
TOPIC STARTER
Yes Nov 17, 2010

Alexandre,

Yes, you will be able to restore initial order of segments. Just choose "No sorting" option when you will finish your work with numbers.

And I agree with you statement about complete mess with numbers translation. Idea to keep an eye on numbers seems to be interesting, but it should be optional, not obligatory.
Because present situation is not predictable. Sometimes autotranslatables work sometimes not. Sometimes TM offers good translations. But the most reliable way is manual copy and edit. Still.

We had some discussions on Russian translators forum and decision is not clear for my Russian colleagues too.


Direct link Reply with quote
 

Alexandre Oberlin  Identity Verified
France
Local time: 18:09
English to French
+ ...
I found a helpful key : F4 Nov 18, 2010

Hi,

Michael Popov wrote:
Yes, you will be able to restore initial order of segments. Just choose "No sorting" option when you will finish your work with numbers.


Thanks Michael for your valuable tip.

I could alleviate part of the bore with big numbers using this command:
Translation -> Insert Assembled Hit (shortcut F4)
It comes handy if you've got big numbers and/or decimal numbers. It will adequately localize the decimal point and the thousands separator. It might of course insert unwanted stuff you'll have to trim, but I found it safe for number strings, and of course quicker than manually editing each figure.

The main annoyance remains those series of segments consisting of simple integers or section titles (e.g. 1.) which all have a TU I don't know why. You must be extra careful because sometimes the numbers don't match! Maybe it is related to the fact that I initialized the TM in TagEditor from previous TTX files and then exported to TMX for retrieval in MemoQ. After that I worked in MemoQ with INX files. Probably I should have retrieved the TM by opening directly the TTX files in MemoQ, but I saw this option deprecated somewhere.

Cheers,

Alexandre


Direct link Reply with quote
 

Alexandre Oberlin  Identity Verified
France
Local time: 18:09
English to French
+ ...
I think I spotted the "feature" Nov 18, 2010

Alexandre Oberlin wrote:
The main annoyance remains those series of segments consisting of simple integers or section titles (e.g. 1.) which all have a TU I don't know why.

I found the following at http://www.translationtribulations.com/2009/12/translating-trados-ttx-files-with-memoq.html It's about importing TTX files into MemoQ:
Selecting the option to import unsegmented content (circled in red with a red arrow pointing to it) will cause numbers and dates, which are usually skipped by Trados, to be imported into the MemoQ project. I don't know of any other software that will do this currently. This is very helpful for technical or financial documents with tables of numbers to be corrected. It is not easy to find all this content in the TagEditor environment, so in this regard, this aspect of quality assurance is a lot easier for a Trados project if it is done in MemoQ.

I don't think I had such a choice when importing the TMX I created in Trados, but obviously unsegmented content was imported.

Cheers,

Alexandre


Direct link Reply with quote
 
Julio A. Juncal  Identity Verified
United States
Local time: 12:09
English to Spanish
+ ...
Need code for autotranslation of billions Oct 12, 2012

I translate from English into Spanish and I frequently have to translate numbers like these:

$45.9 billion

which in Spanish would become "$45.900 millones".

How would you write the code for this autotranslatable rule?

Your expert help would be appreciated.

Julio A. Juncal


Direct link Reply with quote
 

Dominique Pivard  Identity Verified
Local time: 19:09
Finnish to French
A few additional examples Oct 12, 2012

Julio A. Juncal wrote:
I translate from English into Spanish and I frequently have to translate numbers like these:
$45.9 billion
which in Spanish would become "$45.900 millones".
How would you write the code for this autotranslatable rule?

Can you provide a few additional examples of numbers you would want the rule(s) to cover? For instance, would 999.9 billion be the largest number to cover? Would 0.5 billion need to be covered (500 millones?)? would you always have one decimal in English, or perhaps none ($45 billion), or two ($45.25 billion)?


Direct link Reply with quote
 

Kathleen Misson  Identity Verified
Spain
Local time: 18:09
Member (2009)
Spanish to English
Incorrectly formatted numbers in source Jan 12, 2013

Hello,

I think this may be a rule I need to set up somewhere but I am not sure how to go about it.

When a number is formatted correctly in the original (Spanish to English in this case)

e.g.
7.801,21

It is correctly translated
7,801.21

However, if there is a slight formatting error (as is ofen the case with scanned documents), the program seems to select the nearest match from the memory.

e.g.
7.801.21
7 801.21
7 801 21

translated to
953,21.03

Not even close.

How do I create these rules?

Thank you in advance.


Direct link Reply with quote
 


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


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

Automatic "translation" of numbers

Advanced search






memoQ translator pro
Kilgray's memoQ is the world's fastest developing integrated localization & translation environment rendering you more productive and efficient.

With our advanced file filters, unlimited language and advanced file support, memoQ translator pro has been designed for translators and reviewers who work on their own, with other translators or in team-based translation projects.

More info »
Across v6.3
Translation Toolkit and Sales Potential under One Roof

Apart from features that enable you to translate more efficiently, the new Across Translator Edition v6.3 comprises your crossMarket membership. The new online network for Across users assists you in exploring new sales potential and generating revenue.

More info »



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