Computer-assisted translation programs, known as CAT Tools, help the translator in his or her task by making available different features that facilitate translation. For instance, translation memories, glossaries, concordance, and QA. It must be clear that CAT Tools are very different from automatic translation tools, which only require entering the text, that will be translated without any human involvement.

Using these functions that CAT tools provide is very important when it comes to technical translation services, where keeping consistency in terminology is key. When the translation is done, the CAT tool exports the translation into the source format. The main commercial CAT tools in the market are, to name a few, SDL Trados Studio, MemoQ, DejaVu, Wordfast Studio, Cafetran, and a few others. Most of the CAT tools are only Windows-based, but Cafetran and Omega can work in Windows, Macintosh, and Linux.

Open source CAT Tools

There are also open-source, i.e. free, CAT Tools like Omega (a cross-platform desktop tool), BasicCAT, MateCAT (an online tool), SmartCat (an online tool), and a few others.

Main components of CAT tools

Translation Memories

The idea behind Translation Memories, also known as TMs, is simple: they are databases that allow you to reuse and leverage previously completed translations so that they can be applied to new translation projects. Therefore, when using a TM, and dealing with a previously translated sentence, the translation program shows the translator the existing translation in the database, so that it is not necessary to translate it again, which would most likely cause an inconsistency.

However, if the sentence to be translated is not identical to any sentence already existing in the TM, the software searches the database for a similar sentence using an algorithm that allows quantifying the degree of similarity. In other words, the phrase to be translated could be 100% identical (100% match), or 95%, 80%, 75% similar (fuzzy match), or it could be different (No match).

Normally, the translator sets the minimum match percentage that he/she wants to use in his/her translation. For example, you could set the TM to show matches at least 80%. If the match is lower, the TM will not show any translation from its database to the translator.


Whenever we do a technical English to Spanish translation, we create a glossary. A glossary is a collection of definitions or explanations of words that deal with a particular subject and are arranged in alphabetical order. Glossaries may be limited to having two basic fields, the source language, and target language, but may also have additional fields for other languages, fields for definitions of the term, explanatory notes, source, subject, client, etc. Depending on the translation tool, glossaries can highlight the words in the sentence to be translated that are contained in the glossary, or they can simply display them in a window for the translator to apply.

A typical translation program, such as SDL Trados, MemoQ, DejaVu, or Cafetran, allows the creation and application of TMs and glossaries. However, both TMs and glossaries in different CAT tools are not interchangeable.  That is, they are not compatible.

To enable the exchange of TMs, there is an exchange format called TMX, which can be imported from any CAT tool. In addition, some translation tools use TMX directly, and those that use a proprietary format can export their TMs to TMX format. In the case of glossaries, the most common exchange format is MS Excel. Several translation programs allow importing glossaries created in Excel. Those who cannot do so must use additional programs to import from Excel to their glossaries and to export their glossaries to Excel.


An additional feature of TMs is that they allow searching a given word or phrase in the database (TM) to maintain concordance. For example, it could be that the translator finds a word or phrase that he/she remembers having translated before, but does not remember the exact translation. In this case, simply select the word or phrase and press a certain key combination to display the Concordance window where all the sentences containing the word or phrase will appear.

The idea of concordance is to maintain a certain consistency throughout the translation for a given client.


Almost all translation software has built-in translation quality control functions. These functions allow checking: Spelling, translation consistency, terms consistency, tags, double spaces, trailing spaces, numbers format, end punctuation, double words, etc., etc. There is also external QA software, like QA Distiller, Verifika, Xbench, and a few others, which allow quality control and work with bilingual translation files done in the main CAT tools. These external QA tools are much more efficient than the built-in QA function process of CAT tools. They are faster, allow to enter fixes in batch, have better capabilities of search, etc.

Machine Translation (MT)

Machine Translation (MT) or automated translation is a process when a user inputs text in one language and the computer software translates it into a different language, and without human involvement. The main benefit of MT is that translates very quickly and gets the general meaning of the source text. However, for a quality translation, a human translator must edit the output of the MT. Currently, most CAT tools have built-in functionality to work with MT. Among the main MT providers, we have Google Translator, Bing Translator, MyMemory, DeepL, etc.

Additional Notes

In our case, that we offer technical English to Spanish translations, it´s usual to create a glossary in the first place, for the client´s approval, before tackling the translation. This way, we can be sure we are using terminology the client is happy with. Similarly, the client could want to avoid some given technical words in the translation. This is handled by creating a special glossary called blacklist, that warns the translator if he/she tries to use one word present in the blacklist.