Have You Heard of Augmented Translation?

Have You Heard of Augmented Translation?


The days when the main tools available to translators were pen, paper, and a dictionary are long gone. The days of using translation memory as the main productivity resource also seem to be nearing their end. With so many different tools and technologies available on the market, let’s not forget the most important part of the translation process – the human.

Progress in technology means that translators have now gained access to an unprecedented array of aids such as translation memory, terminology management tools, briefing materials, visuals, context information, as well as machine translation and many others.

This has seen a new term being coined to describe the phenomenon. As Eggers, Schatsky and Viechnicki explain in their 2017 whitepaper, “An augmented approach to translation increases productivity and quality while leaving the translator in control of the creative process and responsible for aesthetic judgments.” Sometimes the same concept is also called “intelligence augmentation.”

At TranslateMedia, this translates into the integrated Online Editor interface where translators carry out their work. They have access to several modules that enhance the translation process, as illustrated in the graphic below, but they are still the ones who remain at the very core of the process and stay in control.

A graph explaining how the concept of augmented translation is integrated into TranslateMedia systems

Not directly involved in the process of translating, hence placed on the peripheries of the graph, is the client who is in need of the localization service. The client typically goes to to the Account Manager with their requirements, who then sets the wheels of the translation process in motion. Once the project is started, designated linguists take care of the translation in TranslateMedia’s online-based CAT tool, the Online Editor. Within the tool, the linguists have access to:

  • Terminology Management module, where they see the mandated terms relevant to the text they are translating but also have the opportunity to add their own entries for future reference;
  • Automated Content Enrichment module, which typically means being able to see the visual context corresponding to the text being translated, metadata such as information whether a given segment is a call to action or a button text, etc., but also other information such as reference files and briefing instructions;
  • Adaptive Neural Machine Translation module, available in certain languages, where linguists can choose between different machine translation suggestions and insert them into the segments for further editing. This is designed to provide extra resources to the linguists so that they can increase the speed with which they work;
  • Translation Memory module, where linguists see any matches coming from the translation memory relevant to the text being translated. This also allows for concordance search, in other words looking up how a certain term was previously translated together with the context.

All of the above modules augment and complement linguists’ mental processes involved in translating.

The term “augmented translation” is not entirely new. It has been coined by Common Sense Advisory (CSA), a research company specializing in the area of translation and localization. CSA has drawn inspiration from “augmented reality”, meaning phone, tablet, headset applications overlaying images of the world with additional information like sensory data. Pokémon Go comes to mind as a good example, but the same approach has also gained importance in industrial applications.

In the same way, augmented translation leverages the technology available to allow translators to increase their efficiency and gain more control.

The idea is for all the individual components to “talk” to each other but also learn from the linguist. Machine translation, for instance, uses translator’s corrections to learn new translation patterns and therefore improve over time.

In practical terms, what it means for the translation industry is that companies which embrace the idea of augmented translation and incorporate it in their daily operations, will see a considerable increase in performance that will improve further over time as the systems learn from individual linguists, project managers and customers. 

It is an exciting time to be part of the translation industry, both for language service providers and linguists.

Written by Kasia Kosmaczewska
Kasia Kosmaczewska
Kasia Kosmaczewska is Machine Translation Programme Manager at TranslateMedia. She has extensive linguistic experience and a keen interest in machine learning. She spends her free time reading about socio-politics, practicing pilates and travelling.

Related posts