Businesses that think translation can be fully automated are being extremely shortsighted.
While there have been moves in recent months to bring a new series of tools to market – both Skype and Twitter have hinted at the future potential of machine translation (MT) – the fact remains that professional linguists will always be needed.
We've already outlined a number of reasons why MT is not always suitable, including the fact that language is a subjective method of communication that is inherently complex, so creating a top-quality translation takes expertise and time.
Don't get us wrong, we know MT is a useful tool, especially when companies have a huge amount of content that needs to be translated in a short period of time. However, it will never be a direct replacement for a qualified sector specialist.
Meeting customer expectations
One of the major changes over the past few years has been the increase in customer expectations. In today's technologically-driven world, producing substandard content for the web is simply not acceptable.
For example, YouTube is now offering courses on how to create the most innovative and engaging video possible. Gone are the days when a father could simply take a video of his two kids doing something cute and become an internet sensation with over 14 million views.
This applies to translation too, as consumers are constantly striving for perfection. They do not want to read a product description from another country that leaves them puzzled. If companies want to create an authentic sales journey – and make these people brand advocates along the way – they have to offer a seamless experience regardless of the language.
The future is human
Earlier this year, Google purchased World Lens, an app that translates foreign languages in real time using the iPhone and Android smartphone built-in camera. When combined with Google Translate, it means the tech giant possesses an impressive toolkit for amatuer translators.
But, as with the aforementioned Translate, it is not perfect. The app is great for getting the general point across, but this is only good enough for businesses up to a point. While MT apps can be used by a person looking to find out what is on the menu at a restaurant, the technology is simply not sophisticated enough for pharmaceutical firms looking to translate data sheets and user guides.
Indeed, the Technical University of Berlin has pointed out that MT really falls down when it comes to humor, which is a legitimate form of advertising for businesses of all sizes. Researchers at the institution think that MT tools need to be to able to analyze "the full syntactic chart of the source sentence" in order to be useful.
The main reasons for choosing MT – it's fast, cheap and broadly accurate – mean it is always going to be a popular piece of technology. But the bottom line is that no computer will ever be able to understand the complexities of language like a skilled human.
Posted by Machines will never be able to completely replace professional linguists in the localization process.
