What is Translation Memory?
Translation memory, commonly referred to as ‘TM’, is a database that stores previously translated text. Most commonly used in conjunction with computer aided translation (CAT) tools, such as Trados or DejàVu, it is utilised during the translation process to help with new similar translation content.
Nowadays, a vast amount of translators and translation agencies use CAT tools and TMs as part of their daily work. Translation agencies are able to develop client-specific translation memories in order to produce consistent and cost-effective translations, even when different individual translators are used.
A translation memory consists of text segments (paragraphs or sentences, but not single words) in a source language and their translations into one or more target languages. Together, these pairs of segments form so-called translation units.
Although TMs can be employed in a variety of translation scenarios, they are at their most beneficial when used with highly repetitive texts. They are also particularly helpful when translating newer versions of previously translated documents. In this situation only new content would need to be translated as the TM will identify previously translated content. This can represent a considerable time and cost saving as well as aiding in the production of consistent translations.
Translation memories are extremely advantageous in numerous ways:
Consistency
TMs help to ensure consistency through the presentation of previously translated content. This is particularly important if a number of translators work for the same client or on the same project over time.
