Technical translations and how to post-edit them like a pro

As the translator's role continues to evolve into that of post-editor, efficient strategies for correcting Neural Machine Translation output have never been more important.

Text by Michael Schroeder


Image: © Lamaip/

There is no doubt that the traditional role of translators is changing with automated machine translation driving the revolution. Whereas in the not-too-distant past, computer-generated translations were dubious at best, the improved accuracy of Neural Machine Translation (NMT) has already significantly changed the entire localization industry. This can be seen in the increasing prevalence of multilingual Internet chatbots and on-the-fly presentation captioning. At the same time, localization professionals benefit from higher-quality translations that often just need a bit of touching up to make them suitable for the general public as well as existing or prospective customers.

Despite its recent advances, NMT is still limited by the input it receives: It is not (yet) possible, for example, to effectively "transcreate" an advertisement from a target group in one country to the same ...