Translation productivity and MT

Translation productivity and MT

Never before has there been more content to translate. Not only because in the present-day content is being produced on an unprecedented scale but also because businesses are becoming more and more aware of just how crucial it is to communicate with their customers in their native language. As a result, they are more eager to invest in localisation. But how can translators cope with this deluge without compromising the quality of their work?

Why translation productivity matters

Any translator will have their own individual working pace. While that individual working speed can be determined by a multitude of factors ranging from external, such as the complexity of the source text, to internal, such as how well rested the linguist is, the technology supporting the translation process plays a crucial role too.

The most popular measure of linguist’s productivity applied across the industry is the number of words they can translate in an hour. As reported by TAUS, an average translator translates around 400 words in an hour without the help of any technology.

It makes a lot of sense for linguists to track their productivity. Knowing one’s speed can help in being more accurate in estimating completion deadlines and also charging clients more accurately based on the amount of time spent.

Sand running through the bulbs of an hourglass measuring the passing time in a countdown to a deadline, on a dark background with copy space.

Tracking productivity is crucial to identifying areas of the translation process that can be optimised further.

As the demand for linguistic services increases, not always proportionately to the timeframe in which they need to be provided, linguists are likely to find themselves under pressure. Fortunately, there are a number of means which can help them manage their work.

Translation productivity and technology

Over the years, the language industry has developed an impressive variety of tools that linguists can use to increase their working speed. By far, the one that has had the most groundbreaking impact is translation memory. Its usage has become so common that it is hard to imagine working without it.

RELATED: Have You Heard Of Augmented Translation?

While it is clear that translating faster can lead to time economies and financial benefits, it cannot happen to the detriment of the quality of the final translation. The aim is not to simply be quicker but to work smarter. In other words, do a great job in less time.

In the last few years, translation memory, a technology that is well-established and widely adopted, has started being complemented with its much younger sister – machine translation – to push time-efficiency even further.

Leather tool belt with construction tooling on wooden board maintenance concept.

Technology provides translators with a handy set of tools allowing them to streamline more menial tasks and focus on the ones requiring skill and creativity.

It is not uncommon to think of machine translation input as TM matches, only coming from a different source. When enabled in CAT tools, machine translation input is usually provided on segment level and displayed in a very similar way to translation memory matches to make it easier for linguists to work with it. MT input can be quickly inserted into the active segment using a mouse or keyboard shortcuts and then edited as required.

Machine translation of good quality, coming from a secure source that respects data privacy, can make a significant difference in handling those segments where no TM matches are available, which is exactly where the traditional translation memory offers limited possibilities.

In response to growing demand, CAT tool providers are rushing to incorporate machine translation via plugins, giving linguists the opportunity to seamlessly leverage machine translation in order to speed up their work.

How machine translation can make a difference

Although machine translation will not be helpful in every single scenario (think working with highly creative texts), it offers help where there was previously no help available at all. In ‘no translation memory match’ scenarios, it provides a good base for the final polished translation and can also offer some terminology inspiration.

Particularly in situations where timelines are very short, machine translation can do all the heavy lifting while the linguist can focus on polishing the automated output without having to spend time and energy on translating the content from scratch.

A group of young people hold strong hands. Sign of trust and teamwork.

In the translation industry, linguists and technology go hand in hand.

Machine translation can prove particularly helpful in segments containing well-structured, repetitive text with no ambiguous terminology, such as user manual instructions, standardised legal phrases or strings of numbers and dates. This is where the biggest time savings can be made.

In certain scenarios, using machine translation has been known to double linguist’s working speed, slashing the turnaround time in half and therefore taking away a large chunk of time-pressure from the translator.

If you are unsure how to best leverage machine translation in your everyday work, please contact your Project Manager for some tips.

Related posts

Subscribe to our newsletter