Most translation companies are quick to denounce machine-based translation services like Google Translate. They state that these systems and programs fail in comparison to a human translator, both in accuracy and fluidity.
We are no different. No, at Lingosphere we too believe and know that human translations will always outperform a computer-based translation. Makes sense, doesn’t it? Language is a human-specific trait that is constantly growing and evolving. The ability to adapt a translation based on certain elements of a target audience can only be mastered by a human.
Regardless, we also recognize that Google Translate has its place in translations, and that many people look to this application, perhaps (and hopefully) not for important or expansive projects, but for a quick “how do I” solution.
As such, it’s important for you to know how Google Translate works, so that you understand where your translations are coming from.
At the time of this writing, Google Translate supports 80 languages. At the rate it’s going, by the time you finish reading this sentence, that number might have jumped to 100.
Languages from Swahili to Hindi are all covered by Google Translate, but how?
Putting the human in “computer”
Google knows that language is human-based, which is why that even though a lot of computer work goes into the translation, it all comes down to us humans.
With Google Translate, the program scours through millions of documents, in the blink of an eye, to come up with the perfect translation for your request. The process is called “statistical machine translation. These patterns come from documents that are translated by actual human translators, but the translations themselves, that you see on our screen, come from computers. As such, the translation may not be 100% accurate – as Google admits – but it’s pretty darn close.
The more documents that are translated by humans for Google Translate, the more likely you’ll find an accurate translation on your computer screen. That’s why popular languages like Spanish might have a higher accuracy rate than a less popular language, such as Igbo.
Get involved in Google Translate’s accuracy
Chances are that at some time in your life you’ll find a translation in Google that is wrong. You have several options here. First, you can check out the alternative results, which are typically available. Remember, Google spits out what it thinks is the most accurate translation for you, but there are many other possibilities out there.
Second, you could improve the translation quality by using Google’s Translator Toolkit, or upload your translation memories into the Toolkit to add to the library of translations.
One thing Google Translate likes to state is that its power and accuracy is only getting better with each and every day, as more and more documents are being added to their database.
So why not use Google Translate?
Unless you’re looking for directions or how to say hi to a stranger, Google Translate should not be your only resource for translations. The lack of 100% accuracy is enough to scare anyone away from using it for professional purposes.
Take, for example, this “alternative results” option you have, in case your translation seems a bit off. The question you might have asked already is – “How do I know that my translation is wrong?” If you’re translating text, chances are you don’t know the language to begin with (or else why would you need Google Translate’s help?). Thus, there’s no way for you to tell if your translation is accurate or not, unless you have the expertise of a translator.
Here at Lingosphere, we’re constantly seeking out and embracing new and advanced methods to offer our clients accurate translations. We encourage you to use Google Translate to open your horizons to new worlds and languages.
However, when it comes to professional translations to convey your message to an international audience, you simply can’t outperform our human translators.
Photo Credit by: http://www.flickr.com/photos/augie/