Start a new topic

How to setup a tri-lingual glossary

I've just bought a license (still waiting for the code, though) and I am going to bombard this forum with questions. First thing first, I need to create a tri-lingual glossary for the following language pairs:

Italian-Japanese (both directions)
English to Italian
English to Japanese

When creating a new glossay, the window shown in the attached picture opens in which lowest part all working languages should be specified, but I don't know how exactly. Therefore I am asking, please: to which number in the image should I assign the three languages above (Italian, Japanese and English)  so as to cover all possible combinations?


Many thanks in advance

Mario: I need to create a tri-lingual glossary

Silly question: How are you going to use it. All languages at the same time? What do you want to add to the glossary while you're at the job.

U used to do EU BS, from both English and German to Dutch. I used the SQLite database ("Total Recall) to search the "missing" language. But that's all I could do, search.

H (aka Meta Arkadia)

H: How are you going to use it. All languages at the same time?

I have been looking for this functionality for two main reasons: the ability to cross-check terms in different  languages, in particular with works which source language can be English or Japanese (different files belonging to the same project), and for self-training purposes. This functionality is currently offered by SDL Studio; and also by MemoQ when I tried it several years ago. 

I understand this is not something that most translators would be intersted in, but I found it helpful.


Mario: different files belonging to the same project

That's exactly how I used it (in CafeTran) for my EU projects. I don't see, however, how I could easily add to the "missing" language. So if I translate EN>NL, but I want to check the term in German, I can do it, but I can't add info to it.



Hi Mario,

I have never used a multilingual glossary in Cafetran myself, but it can be easily done with CafeTran's .txt glossaries. CafeTran has two main terminology formats: 

1. TMX files + 

2. Tab-delimited txt files (which are currently called "Glossaries" in CT lingo, unless it has changed again).


The format of a simple multilingual .txt Glossary would be:

#en-US TAB #es-ES TAB #de-DE TAB #Subject TAB #Client TAB etc.

These glossaries are very easy to manipulate in a good text editor, and particularly so in a good "CSV editor". I myself use: , but I am on Windows. There must be something similar on Mac. 

Hans mentioned CafeTran's "Total Recall" system, which is quite a different beast, and not where you should begin as a new user. That's a special system based on SQLite (and other underlying db formats) to access very large databases of bilingual data (TBs or TMs).

If I have a moment I could help you set up a basic multilingual txt glossary.

Re yr question, I think your 3 languages should be assigned to numbers 1, 2 and 3, but Igor would have to confirm this, or someone else who has used a multilingual Glossary.


1 person likes this

Michael: can be easily done with CafeTran's .txt glossaries

You know I'm not particularly a fan of tab del .txt glossaries, but this could be the solution IF:

  • The 3rd (non-project) language is shown, other than in the edit glossary pane
  • The 3rd (non-project) language can be searched for and edited
You can have multilingual TM for Fragments TMX files as well, but I don't think they will meet the criteria above either.

.... not where you should begin as a new user.

I disagree. Setting up an SQLite database in CafeTran ("Total Recall") is perfectly easy. Just as easy as setting up your other resources. Just import your TMX TMs or tab dels. In fact, Igor recommends to start using them from your very first project (though I disagree with that).


If you wish to create a new multilingual dictionary, just make sure the three (or more) languages are selected in Preferences >  Definitions tab > Language column names (from left to right). Then, CafeTran will always create a glossary file with the columns for those languages. The current (main) language pair of the multilingual glossary is selected automatically based on your current project's language pair.

IK: The current (main) language pair of the multilingual glossary is selected automatically based on your current project's language pair.

Yes, but my understanding was/is, that Mario wants to be able to look up in the third language. Like, he's doing an English to Italian job, but wants to know the meaning of a word/term in the SL or TL in Japanese.


Then all he needs to do is to specify the language of the look-up via the right-click at the glossary pane and choosing it in the Search submenu.

IK: ...specify the language of the look-up via the right-click at the glossary pane and choosing it in the Search submenu.

Great! That's the kind of solution I was waiting for, and I bet Mario too.

Can you provide a screenshot?


Actually, the user doesn't need to do anything as by default CT searches all defined language columns. He can narrow down the search as shown below:

It seems that I'm going to have fun with my glorious tri-lingual glossary. For the moment I'll try to complete my new 3-column Excel file as soon as possible (I'm already at letter "Q"), then I will see whether I will be actually able to get what I desired most from CTE.

Thanks Igor and Hans

But, if my understanding is correct so far, it will not be possible to display the third language simultaneously like, for instance, in SDL Studio and MemoQ? That is, two more clicks are necessary? 

I new I was going to have fun with my trilingual glossary! Here is what happend:

  • I have 3-column Excel file (.xlsx): IT, EN and JP (some EN and/or JP translation are still missing, for I plan to insert them later on while translating)
  • The newly created IT glossary (empty for the moment) is set for the three languages above (it-IT, en-GB and jp-JP, plus other fields which I am going to leave empty for now.
  • When I selected "Import MS Excel glossary" and then the Excel file above, nothing happened visually, leaving me in the doubt that nothing is working. After a while CT said that the Excel file has been imported, but when I opened the glossay I noticed that CT has imported correctly only the IT lemmas, but for the en-GP and the jp-JP it just imported what it found first. For instance, if for one lemma there is only the Japanese translation available, CT imports this translation and puts iy in the place of the EN translation instead of leaving the en-GP field empty.

In fact, during the Excel file import process CT is not asking to what glossary field the three Excel columns should be mapped, which I believe is fundamental to put each language translation in its proper place.

Hence my question: can CT really import a tri-lingual Excel file into a glossary?


Login to post a comment