When professional translators (preferably with subject-matter expertise in a given field) are translating a document into their target language, it’s an industry best practice for them to employ a translation memory system (TMS). This is especially the case for all large-scale localization projects, which often involve collaboration between two or more translators. Translation memory systems allow translators to effectively leverage all previously translated material, and they also promote linguistic harmony between an array of interconnected documents.
There are two components to most translation memory environments available for professional use: a segment (or sentence) repository and a glossary. The sentence repository function (which is the “memory” aspect of the system) is essentially a database of linked source and target segments (including sentences and paragraphs, as well as smaller sub-sentential elements). When translation memory software detects a match between the current source document and previously translated material, it prompts translators to reuse the target language content. The most powerful translation memory systems utilized by professionals in the localization industry today employ fuzzy matching algorithms that not only retrieve previously translated sentences and paragraphs in their entirety, but also provide linguists with translated segments that are similar to the source text, with any differences highlighted in the operational window. It is the translator’s job to leverage any applicable portions of the already translated content, and then manually translate the segments of text that differ from those present in the translation memory.
The second component of translation memory systems—the glossary component—is not merely a sentence repository, but rather a word depository. This is the function that deals directly with terminology management. As stated, in instances where a TMS does not supply any previously translated material, the linguists working on the project must manually translate all new content. When you integrate an up-to-date multilingual glossary with the translation memory system, however, the software will automatically propose approved translations for individual words or terms.
There are several benefits that translators derive from the glossary function in translation memory systems. First and foremost, the glossary lookup function is fully automatic—it supplies translated terms on an individual basis, along with a suite of applicable attributes, like the term’s definition, its part of speech, and its status as a preferred or deprecated term, etc. This information is automatically supplied to translators so they don’t have to waste time or effort thinking about how to translate a particular term, nor do they have to research the word in question, as all of the term’s definitive characteristics are displayed right in the translation environment. With the time spent on researching unfamiliar terminology reduced to a minimum, a multilingual glossary can significantly boost the level of a translator’s productivity.
More importantly, however, a multilingual glossary enables translators to use translated terms that have been pre-approved by the client. By incorporating authorized translations of a given set of terms into their work, translators can produce content that is consistent across languages, documents and between previous versions of product literature. This agreement between translated documents has measurably positive effects on an organization’s branding overseas, and enables translators to not only leverage content (which saves money), but also enables them to leverage translation quality (which improves product usability).