What is machine translation and how does it work?

If you’ve ever used Google Translate, you’ve used a machine translation tool. Machine translation is instantaneous – you input the source language text, press a button, and within a matter of moments, the text is returned to you in the target language. It can even be used to translate an entire webpage.

Some machine translation systems are built on bilingual corpora, or sets of documents that have already been translated. By referencing existing translations, the tool is able to identify a possible translation for your request. This is known as a statistical machine translation tool. Other systems, known as rule-based machine translation, rely on analysis of grammatical rules to generate a translation.

What is human translation?

Human translation is just what it sounds like – translation performed by people. A human translator will translate each segment of the source text at a time, taking into consideration register, context, available space, style guides and glossaries, and more factors, which we’ll discuss below.

Many human translators do use software to aid in the translation process, increasing both productivity and accuracy. This type of software is known as a computer-aided translation tool, or CAT tool. A CAT tool can recognize when a translator has previously translated a similar sentence, and can suggest a possible translation for the translator to review and implement as appropriate. It can also remind translators of glossary terms, and help them to easily format a translated document.

CAT tools are not the same as machine translation tools – they still require a human translator to review any existing translations, and input new ones wherever there is an untranslated segment.

Why should I choose human translation?

There are many reasons to select professional human translation over machine translation.
The most basic reason is simply to avoid the type of errors that often arise when a machine translation tool isn’t able to recognize and cope with the complexities of human language.