UK +44 (0)207 078 4350
View homepage in:

There can be confusion between the use of translation memory tools by human translators and what has become known as machine translation.  The team at Technical Translations are all human translators, but here we examine the role for machine translation and where it should never be used.

Can Machine Translation Really Replace Humans?

Machine translation has been improving over time, although Technical Translations still get clients who have had an uncomfortable brush with its results coming to us to put right mistakes made by machine translation. We all know that large enterprises such as Google are putting a lot of money into improving their free machine translation service, but how effective is machine translation and how do you know whether your translation vendor is using it or not?

How Effective Is Machine Translation?

We at Technical Translations would only recommend our clients use a tool like Google Translate (and we do) to enable them to understand the gist of a communication they have received in a foreign language, for instance, so that they can decide whether it is something they’d like to have translated by a qualified human translator. The reason for this? Well there’s a lot that machines miss.

Google’s online translation tool, for example, works just like its search engine does, using a database of texts from the internet (examples of word variants and phrase matches between different languages) and uses algorithms to enable it to pick the best match to the words and phrases in the source text.  However, it isn’t perfect and Google point this out to its users in their blog:

When Google Translate generates a translation, it looks for patterns in hundreds of millions of documents to help decide on the best translation for you. By detecting patterns in documents that have already been translated by human translators, Google Translate can make intelligent guesses as to what an appropriate translation should be. This process of seeking patterns in large amounts of text is called “statistical machine translation”. Since the translations are generated by machines, not all translation will be perfect. The more human-translated documents that Google Translate can analyse in a specific language, the better the translation quality will be. This is why translation accuracy will sometimes vary across languages.

Machine translation is a very useful tool, but it has its limits and if accuracy is what you’re looking for, a professional translator can still deliver the best results.

How Do I Know If My Vendor Is Using Machine Translation Tools?

You don’t, unless they advertise the fact in their marketing material. However, if you’ve been having trouble with the quality of the output of your translation provider and their service is unbelievably cheap, it could be that they’re using machine translation to keep the cost down. In the aftermath of the recent recession everyone is looking for ways to cut costs and translation providers who do use machine translation for the bulk of their work often undercut those using trained professionals.

However, because machine translation tools use formal and systematic rules, there is a high risk of ambiguity of meaning and of confusion. It is generally agreed in the translation industry that while machine translation tools are handy for translating high volumes, where only a gist is required for instance, a professional translator is irreplaceable where accuracy of content and meaning are essential.


Read More