How does translation memory and translation glossary work, and what is the difference?

Translation Memory

A translation memory (TM) is a database that stores "segments", which can be sentences, paragraphs or sentence-like units (headings, titles or elements in a list) that have previously been translated. When we encounter text that exactly matches existing memory segments, we will use memory entry instead of having the text re-translated. We match TM entries on a per-sentence level or an entire order level. We do not combine multiple sentences together within a larger order, nor do we do fuzzy-matches at this time.


Translation Glossary

A translation glossary allows you to specify how you want specific words or phrases translated. For example: “automobile”, “car” and “vehicle” are all accurate ways to call a vehicle. By default, the correct word to use for the translation is up to the translator’s discretion. By adding in glossary entries, you can tell the translator that you want the word “automobile” used in place of “car” and “vehicle”.  We match glossary entries on a single word or a phrase, as long as the text we are replacing exactly matches the glossary entry.


What’s the Difference?

Translation memory is matched on a per-order or per-sentence level. It tends to be larger entries that are used increase accuracy and seed up the translation process. Glossary entries tend to be much shorter and are used to specify how you want specific words and phrases translated.


How Does TM Effect The Order Cost?

If you submit an order that has been previously submitted in your account, and completed by us before there is no charge for us to process it again. If only specific sentences are found with the memory, we still put the order through our entire translation process so there is no change in the cost of the order.