Translation memories (TMs) make it easier to translate files as they change multiple times, ensuring consistency and reducing the time and cost of translation. You most commonly work with TM entries on one or more individual translation memory pages.
When you translate an asset or task and save to translation memory, the source and target pairs populate the translation memory. When you translate those assets again, all unchanged source segments will have a 100% target match or an ICE/SPICE match, depending on the segments around them. The source segments that may have changed slightly will have a lower-scoring match that you can access in a translation workbench by clicking the translation memory button for the segment.
In addition, translation memories save time and promote consistency when translating other files. Often, multiple files contain the same or similar text (for example, your company's tagline). A translation memory can ensure that translations of the same (or similar) pieces of text across different files gets translated similarly. This helps promote consistency across all of your content.
The Translation Memories page has the following columns.
To view or work on a particular translation memory, simply click on the name of the TM. This opens the individual TM page, where you can view the TM, work with it in a translation workbench, search for and replace entries, and import and export entries.