What is a Translation Memory?
Before we introduce our Translation Memory FAQs, let’s start with the basics: CAT tools are software which help translators and language reviewers work better and faster. A Translation Memory optimises the process in three ways: improving the quality of translations, reducing the time they take, and reducing the costs associated.
This post is a continuation of our recent article:
What is Translation Memory?
For more general information on CAT tools and/or Translation Memories, start there!
Here are some of our Translation Memory FAQs:
The main thing this software does is ‘remember’ how we translate each segment of text. CAT tools split a source text into manageable units known as ‘segments’. The software then builds databases of equivalent segments in different languages. The databases of these matching segments form a Translation Memory (TM). Therefore, if the same segment appears again in future projects, the Translation Memory (database) will offer the ‘remembered’ translation as a suggestion.
Translation Memory FAQs
What is a segment?
A segment is a unit of text. It could be a sentence, a heading, or a piece of technical vocabulary, amongst others. Translation Memory (TM) works by breaking a text down into segments, and ‘remembering’ how they were translated previously.
How do you handle segments that are repeated? (“Repetitions”)
Once a sentence has been translated, if it is repeated within the text the software will automatically ‘suggest’ the previous translation when it appears the second time. The translator will then check the suggested text — to see if it works in this context, or whether they need to modify or re-translate entirely. For a simple segment, the suggestion will generally be appropriate.
What is a “Match”?
As soon as we have translated one document, all of the segments go into a Translation Memory specific to that particular client. When we work on further documents for the same client, we check any new translation projects against their TM. The software will highlight any sentences which are the same as, or similar to, sentences we translated in previous documents. If the segments are identical, we call this a “100% match”. The translator will then check that the match is appropriate in the context, and, if it is, insert it directly.
What is a “Fuzzy Match”?
A fuzzy match is a match which has less than 100% identification with the new sentence. If a match is ‘fuzzy’ the translator will check it against the sentence to be translated, and either modify it, or reject the suggested match.
How does TM affect quality, speed and price?
Use of a TM improves quality, increases speed and reduces price:
- We can maintain high quality by ensuring consistency across all the translations we undertake for a given company.
- We can speed up the translation process, as our translators don’t need to re-translate segments which have already been translated, and approved.
- And we can offer competitive pricing because QuickSilver will offer you a substantial discount for repeated or matched segments.