5 Localization Trends for 2021

After a year like this one, the healthiest thing to do is look to the future. And what’s that on the horizon? A whole raft of intriguing developments in the translation industry! Read on as Milengo presents the key localization trends you should be aware of when drawing up your strategy for 2021.

1. Diversity in communication

By 2022, videos will account for 82% of all data traffic worldwide. At the same time, thanks to the roaring success of remote working this year, the e-learning market is predicted to keep growing, achieving a worth of around 325 billion US dollars by 2025. These developments will bring with them vast potential for multimedia translation, driven by trends such as microlearning, gamification and virtual reality. With more and more people sending videos to each other and watching them with the sound off at work, the demand for subtitles, text captions etc. has never been higher.

Artificial intelligence will have an important role to play when it comes to localizing multimedia content. One example of this is automated transcription processes, which are used for applications such as the generation of closed-caption subtitles. In addition to this, systems such as Google’s Translatotron are now also able to translate spoken language directly using speech-to-text and text-to-speech technology. In future, companies looking to use video content as one of the cornerstones of their global communication strategy will need a partner who not only understands how translation processes work, but also knows how to handle multimedia localization.

Our New Year’s resolution:

To use cutting-edge AI technology to make global learning a readily available and affordable service in every language

“In times of crisis, translations play an important role in keeping global channels of communication open. In 2021, trends such as AI and automation will trigger another leap forward in terms of innovation in our industry.”

Roman Kotzsch
Milengo CEO

2. New forms of human-machine interaction

Machine translation, or MT for short, is sure to leave its mark on the next decade – of that there is no doubt. With an annual growth rate of 17 percent, current forecasts predict that the market will reach a volume of 1.5 billion US dollars by 2026, driven primarily by sectors such as the automotive, software and finance industries. However, the initial hype that surrounded neural MT when it was first launched in 2016 has since deflated somewhat. Despite the impressive results delivered by deep learning technology, computers have yet to demonstrate the ability to replicate the complex cognitive processes required to subjectively interpret and produce the capricious beast that is human language.

As a result of this stalled development, the idea of a competition between man and machine has given way to a more symbiotic relationship. Market analysts Gartner refer to this concept as augmented intelligence – the notion that, rather than demoting humans to mere dogsbodies in the aid of technological progress, technology should instead be helping us to become better, more intelligent and happier.

Augmented translation presents a number of exciting opportunities for human-machine interaction. One such example is Milengo’s innovative post-editing workflow, Managed MT – a modular, AI-assisted quality assurance process that raises the bar in terms of productivity, especially when processing large volumes of text.

Our New Year’s resolution:

To strike the perfect balance between the linguistic finesse of the human mind and the superior efficiency of machine translation

3. A scarce resource

Legendary science-fiction writer Stanislaw Lem once warned that artificial intelligence could end up going hand-in-hand with the “deterioration of human imagination and intelligence”. This danger is just as real in the translation industry as anywhere else, and it is all too easy for translators to fall into the trap of leaning too heavily on DeepL and other digital tools when they should be using texts to hone their own translation skills instead.

Well-trained translators are in notably short supply – especially in disciplines such as marketing translation, where creativity and linguistic flair are just as essential as terminological and grammatical accuracy. One of the main reasons for this is the rise of machine translation and the increasing wage pressure in the sector, which limit the amount of viable career opportunities available and put young people off pursuing translation as a traditional course of study. As a result of this, competition between companies for the best new translation talent is getting fiercer by the day.

All of this means that there are a few things you need to think about when working with a translation provider. Do they apply strict criteria when selecting their translators? Do they prioritize experienced translators, or those with degrees? Does the agency offer good working conditions, thus preventing the kind of high staff turnover that would have a negative impact on the quality of their translations? There are even a few things you can do yourself to help you get better translations: Providing translators with product training, style guides and terminology glossaries will make their work easier and lay the foundations for a long-lasting, successful partnership.

Our New Year’s resolution:

To develop a new appreciation of the work provided by translators – this shouldn’t be taken for granted!

4. Automation is optimization

Hungarian start-up BeLazy develops middleware that bridges the gap between translation management systems, CAT tools and the various business management systems used by different customers. In doing so, the former memoQ employees and their team have truly tapped into the Zeitgeist, as many sectors begin to implement integration technology in an attempt to upgrade work processes that are slow and susceptible to errors.

The software uses connectors and APIs to network all the components in the translation workflow and transform them into a highly automated system. “Robust connectivity” will be one of the watchwords for 2021 – especially among companies who need to handle large numbers of small, repetitive translation projects that generate a disproportionate amount of admin.

All the same, automation isn’t a mantra to be followed unquestioningly, and many of its purported benefits to the translation industry are still hotly disputed. Can automatic quality assessment systems such as BLEU-Score ensure that machine translations produce satisfactory results? Does it make sense to use AI when recruiting suitable translators for a job? In both of these cases, automated assistance is no substitute for the discerning judgement of human experts.

Our New Year’s resolution:

To develop an integrated translation cycle that enables everyone involved to work together smoothly using tools that were previously incompatible – with no frustration, fiddly and time-consuming tasks, or repetitive copying and pasting.

5. Shifting markets

The Asia-Pacific area is the most dynamic region in the world. In mid-November 2020, 15 states in this region signed the RCEP trade agreement – thus creating the world’s largest free-trade zone. The RCEP reflects the growing confidence of the region and its 2.2 billion inhabitants on the world stage (by way of comparison, the EU has a population of around 447 million), and the massive reduction in customs duties it entails is expected to lead to an expansion in trade and investment. Forecasts predict that China, Japan and South Korea will be the main beneficiaries of the deal, and that it could make European and American export companies less competitive in East and South-East Asia. This could lead to a new boom in the translation industry, especially for providers who can offer services in South-East Asian languages such as Vietnamese, Thai and Indonesian.