A translation memory (TM) is basically a database of previously translated segments for CAT (computer-assisted translation) tools. Different CAT tools use different TM formats, but most of them can be converted. It’s typically one source language and one or more target languages.
TMs work at the sentence level. CAT tools break down source documents into their component sentences or segments. The segment is the smallest reusable chunk of text. Words are not used for this purpose because different contexts require different translations.
As the translator works, the current segment is compared to those in the translation memory, and if it has something very similar, the CAT tool will automatically show this to the translator. Identical source segments are called 100% matches. This means that somebody in the past had already translated that exact segment. There are also 101% and 102% matches, which means that not only the current segment, but also one or both of those before and/or after it are the same as stored in this TM entry. Matches below 100% are called fuzzy matches. These are ranked from 0% to 99%. A 99% match means that the segments differ by at least one character. Matches below 70% are often considered useless and might not show up, depending on the settings.
“Repetitions” are identical segments within one document that have no translation in the translation memory yet. Most CAT tools scan for repetitions before the translator starts working. After the translator is done with the first occurrence of these, all others will get filled in automatically.
Translation memory statistics is a breakdown of how many 100%+ matches, fuzzy matches, repetitions, and new text the file has. Typically, each category has a different price applied to it. For example, a 99% match can be priced at 10% of the default per-word rate, while a 75% match can be priced at 40%. These settings are configurable in most CAT tools. Words “discounted” according to such percentages are sometimes referred to as “weighted” words. The rationale behind this is that a translator works less on matching segments, thus a lowered per-word rate nevertheless results in the same or higher per-hour rate. This is considered a fair deal and is practiced by almost every translation agency and customer nowadays.
Translation memories reduce the price for clients when there is a lot of repetitive content across their past translations while reducing the time required to finish a project.
But most importantly, translation memories ensure the consistency and hence quality of translation. Besides matches as such, you can use your existing translation memories for concordance searches — where you investigate if the translation memory has a certain term was translated before, if the segment as a whole wasn’t.
Note that for even more consistent terminology management it might be better to use glossaries instead of translation memories. We describe glossaries in more detail elsewhere, but here we will just mention that in many cases they ensure even better quality than translation memories, especially when a translation memory is made up of translations of a lot of documents from different clients. Translation memories just won’t “know” what translation your particular client prefers.
We recommend keeping your translation memories and glossaries up-to-date at all times.
There are a lot of translation memory tools out there. These include Wordfast, Trados, Smartcat, Memsource, MemoQ, Across, etc. Some language service providers even make their own translation tools, but these are typically of lower quality and only provide the most basic functions.
Some CAT tools use their own TM formats, for example Trados uses .sdltm. The more or less standard one is .tmx, which is supported by a lot of translation memory tools.
Productivity is today’s number one priority in the translation industry. We all know that we need quality, but we cannot forget that deadlines are out there, and they wait no one.
Smartcat is fine-tuned to provide the highest productivity possible. It does not require installation, is free for unlimited use, and arguably has the easiest and most intuitive UX out there. Once you upload a document, Smartcat splits it in into segments and you can proceed to translating right away. You will see the original text to the left, and enter the translation to the right. In the end, you download the resulting document.
As for the Translation Memory feature, you don’t have to have an existing translation memory when starting to work, Smartcat will create one for you the moment you start a project. If you do have existing translation memories, you can import them. Both .tmx and sdltm file formats are supported.
Smartcat also allows categorizing translation memories by clients or subject areas. Thus, you can have different TMs for different types of content (e.g., software, technical, contracts) and avoid storing all translations in one translation memory, which may result in a big mess that won’t be of much use.
In Smartcat you can also use several translation memories in one project. This is useful in many cases, but particularly when a project cannot be categorized in a single subject area — for instance, when you have one translation memory for software and one for contracts, and the project is about a contract concerning a piece of software. Keep in mind, though, that you can only save translations to one translation memory. So if this is a typical setting for a project, you might want to create a separate translation memory for “software contracts”. We discuss organizing linguistic resources in more detail here.
As already said, Smartcat is free for unlimited use, so you can give it a try and decide whether you like it or not without any commitments.
|Terminology management:||Smartcat||Trados 2017||MemoQ||MemSource||MateCat||XTM||Smartling|
|Moderated terminology base (ability to verify new terms by specialist)||✓||✗||✓||✓||✗||✓||✓|
|Several terminology bases in one project||✓||✓||✓||✓||✓||✓||✓|
|Ability to view the use of the term context||✓||✓||✓||✓||✓||✗||✓|
|Work with terminology without any extra software||✓||✗||✓||✓||✓||✓||✓|
|Ability to add multimedia to the term in glossary||✓||✓||✓||✗||✗||✓||✗|
|Morphological search engine in dictionaries and glossaries||✓||✗||✗||✗||✗||✗||✗|
|Extra settings of glossary structure||✓||✗||✗||✗||✗||✗||✗|
|Number of formats to make terminology base import||2||7||2||2||1||3||1|
|Export terminology base into Excel||✓||✓||✓||✓||✗||✓||✓|
Start boosting your productivity with Translation Memories today.