The most conventional definition of translation quality is that the translated text should be grammatically correct, have correct spelling and punctuation and sound as if it was originally written by a native speaker of the target language. We will refer to all quality assurance tasks performed to ensure this type of quality as linguistic. Obviously most of these tasks require human intervention and are hard to be automated.
However, there is another aspect of quality which we will call formatting. Ensuring such type of quality means not only making certain that the original text formatting was not damaged, but also detecting unnecessary double spaces, double full stops at the end of a sentence, verifying that the same sentences were not accidentally translated differently and that the project glossary was followed, and numerous other tasks which do not require working knowledge of the target language. Usage of translation memories gives rise to the need to ensure that the translator did not apply fuzzy matches as perfect ones, made necessary corrections in the fuzzy matches as well as to make certain that the perfect matches were correct and did not get into the translation memory by mistake. Apparently these tasks require a lot of attention, even thoroughness, and are therefore highly error-prone.
Examples of common translation QA tools:
- ApSIC Xbench (free)
ApSIC Xbench features powerful Quality Assurance features, to check for consistency, numeric values and more.
- D.O.G. ErrorSpy
With ErrorSpy, D.O.G. GmbH has developed the first commercial quality assurance software for translations. ErrorSpy is used to assist the work of the reviser.
Verifika detects formal, consistency, terminology and spelling errors, and features a built-in editor, saving your time to resolve the errors. It also provides smart auto-corrections to save even more time and has a lot of other useful features to help you enjoy QA.
- Yamagata QA Distiller
Yamagata Europe’s QA Distiller is a software application that makes it possible to automatically detect formal errors in translations and translation memories, and enables their quick and easy correction.