Computer-Assisted Translation/Translation Memory (CAT/TM) is a translation software tool that works as a linguistic database to store source text and capture a human translator’s corresponding translated content as they work. The software saves these “segments” of text, such as sentences, sentence-like units, and paragraphs to be re-used in the future, thus improving the efficiency and consistency of human translators.
Segments of text are typically broken down into “No Match” (or “new” text), “Fuzzy Match” (a partial match between source text and a previous translated segment), and “Repetitions” or “100% Match,” where a translation of an entire segment can be re-used completely, without modification.
This model benefits translators by making them more efficient, and it also helps the client because they will usually not pay full price for repetitive content. These savings can be significant, particularly in highly-repetitive texts, such as websites, technical manuals, and software applications.
It is important to remember that TM/CAT tools do not produce “machine translations,” although some TM/CAT tools do possess controlled Machine Translation (MT) functionality, which is used in conjunction with the TM/CAT tool and human translator / editor (this is known as “Human-Assisted Machine Translation” or HAMT). Therefore, a translation performed using a TM/CAT tool is still a “human translation.”
PGLS employs TM/CAT tools on certain applicable projects, although it cannot be applied to certain types of documents, such as scanned PDFs or hard copies, as the software cannot penetrate those types of files.
Please contact us if you would like to learn more about this technology and if it can be applied to your translation project.