These tools (CAT, or Computer Assisted Translations) are centred around the use of translation memories (TM). The principle is that the translated text is split into individual parts, so-called segments (usually sentences), which are then translated by translators. The software stores the segments as pairs of the original language and the matching translations into a database. Whenever an identical or similar text needs to be translated again (whether within a document or at any time in the future), the system identifies this and offers the previous translation as a hint or a suggestion.
CAT tools (among the most widespread are Memsource, Trados, and MemoQ) help provide consistent translations and are especially ideal for translating texts in the following domains:
Another important aspect of CAT tools is terminology (dictionary, glossary), which is not created automatically, but rather systematically by a translator or translators as they work. Glossaries consist of terms in both the source and target languages, so they can be practically described as electronic dictionaries.
A major advantage is that terms are automatically identified in the translated texts, and the corresponding equivalents are provided. Glossaries are useful, especially for the translations of technical texts where terminology is critical, such as in chemistry, where the system identifies words like carbon dioxide and greenhouse effect and provides the best matching translations. Translation memories can also serve as a source of terminology.
Author: Václav Baláček