Translation has been transformed by technology. In just a few decades, the integration of Machine Translation (MT) and Computer-Assisted Translation (CAT) tools into translation work has dramatically increased production speed, improved accuracy, and decreased costs and tedium.


Machine Translation (MT) is an umbrella term for a variety of translation technologies. Commonly available MT like Google Translate will process translations at a level of accuracy that can help users understand and communicate basic concepts. This consumer-grade MT is not usually adequate for creating any type of complete document.

The more sophisticated MT used by LSPs like Skrivanek takes dedicated learning and practice to master. This type of MT is a tool with various memory functions that assist a human translator as they work through a document.

The document to be translated is entered into the programmed MT module. Before translation begins, specific terminology from the client is entered into the program also. The human translator translates the document and during that process they receive suggestions from the MT.

These suggestions come from the „remembered“/stored phrases and terminology from clients and from past jobs. They are offered by the MT to the human translator to select or reject, in order that the translator does not have to re-think how to translate each repeated word or phrase every time it arises.

As you can imagine, use of stored memories of words and phrases fosters consistent word choices and spelling in the final document. This leads to improved accuracy and a signficant amount of time saved. As a result, the price you pay for your translations is significantly reduced.

MT reduces the cost of translation for the client. Purchase of MT technology and the necessary translator training is a serious financial investment for an LSP. But this cost is one that pays off in higher customer satisfaction, more consistent quality, and quicker job turnaround.

MT is accurate enough to be an extremely useful tool for human translators. Its accuracy is greater than just ten years ago, and it continues to improve with advancements such as Neural Machine Translation (NMT). But still, accuracy varies according to the target language as well as the language combination and subject matter. MT cannot be relied upon to produce final translations without human processing and post editing.

The type of documents that benefit most from MT are those with repeated phrases and specialized terminology. Structured documents with repeated phrases such as technical descriptions or standard legal forms are examples.

No, but MT continues to improve as a complex, powerful tool. Together, human translators and MT make an excellent translation team for the production of professional-quality texts.

MT is sometimes used by itself to generate internal company translations of minor documents such as memos, or in other situations where quality can be inconsistent without negative consequences. With the amount of online communication that occurs today, this can be quite helpful.

Yes, absolutely. Our security protocols apply to documents processed using MT.

Ask for a quote

Whatever your translation needs – contact us and we will prepare a custom tailored price quote for you.