Post-editing: A New Skill For Translators


Until not so long ago, post-editing was a cause for controversy, based on the wrong assumption that both machine translation and post-editing were a way to maximize productivity and cut translation costs.

Most people in the translation industry know now that machine translation isn’t a 1-2-3 proposition. In the first place, usually MT only comes into play when huge volumes of data— typically hundreds of thousands of words —need to be translated quickly and repeatedly. In addition, implementing customized machine translation systems requires time as well as, most importantly, a serious financial investment. The choice of a machine translation system depends on many parameters, including the language pairs involved and volume of data to be translated.

Likewise, post-editing a machine-translated text doesn’t simply mean a quick, low-cost look at a text for mistakes. Before post-editing can even begin, the output must be quality controlled. To assist post-editors, skilled companies and LSPs create guidelines based on client expectations; the target audience; the purpose of the text; and many other factors.

Translators can benefit from an unbiased approach to machine translation. Those who take on post-editing assignments need to grasp what machine translation really is. They must differentiate between the various machine translation systems and understand the errors that can occur.

Also, before the assignment even crosses the post-editor’s desk, the post-editing task needs to be defined: the parameters set, the client’s needs and expectations outlined, the intended use of the text—as perishable or publishable—explained, the machine translation output needs to be evaluated.

The TAUS Online Post-Editing Course: A Common Reference For The Translation Industry

Because post-editing is an inherently different skill from revision and since there wasn’t any course available to the translation industry professionals, in 2014 TAUS decided to develop an online post-editing course. The course represents a collaborative industry initiative with the goal of setting post-editing best practices and defining the necessary skills, both for freelancers and translation companies.

The theory part of the course - developed in cooperation with Hunnect and Welocalize – comprises six modules: types of MT, error typology evaluation, pre-editing, controlled language, post-editing as well as some best practices for setting up a machine translation/post-editing project. Each module has a relevant bibliography and a self-assessment test at the end. To complete this first part of the course, participants need on average six hours.

The practical section – created in cooperation with various language partners – makes use of the DQF tool and comprises two exercises: error typology evaluation and productivity measurement. To complete this practical section, participants will need another 5-6 hours. The exercises are language-specific and available in 20 languages, including Japanese.

The TAUS online post-editing course is meant as a gentle introduction to post-editing and can be followed by anyone who has an interest in machine translation, but most specifically by translators, staff of translation agencies, technical writers, translation students and researchers.

Since the launch of the course in March 2014, already over 600 persons have taken or are taking the course. Among them, a group of Japanese translation students who volunteered to translate and post-edit the TAUS website into Japanese, under the expert guide of Mr. Daoud Akira POPAL and Ms. Rie Miyake from the Honyaku Center.

The participants following the TAUS online post-editing course will gain a solid background knowledge of machine translation and post-editing, and develop the necessary skills to provide high-quality post-edited content. Upon completion of the course, they will receive the TAUS Certificate.

Translation agencies and companies looking for experiences and/or TAUS-certified post-editors can find them in the TAUS Post-editors Directory.