Years ago, I remember all of the sites that included those awful automated translation buttons. You’d click the button on a non-English site and it was barely readable. The best test was to translate a paragraph from English to another language… and then back to English to see how different the result was.
Case in point, if I translate the first paragraph from English to Spanish and back again using Google Translate, here’s what the result is:
Years ago, I remember all those buttons sites including horrible machine translation. You click the button on a site other than English and was barely legible. The best proof was to translate a paragraph from English to another language … and then back to English to see how different the outcome was.
In one simple step, you can see the chunk of accuracy and smooth verbiage that’s lost. The limitations of machine translation are the same as they’ve been for years. Machine translation lacks context, the ability to overcome ambiguity, and a lack of experience. The machine isn’t educated with 20+ years in a specific field or topic that has evolved over time. Words aren’t simply translated, they’re interpreted based on the topic and the experience of the writer and reader.
Of course, a human translator won’t fit in your pocket, and they may not always be able to accompany you to that very authentic Thai restaurant or overseas vacation, so here’s what we recommend: When you need immediate results, and they don’t have to be perfect, it’s OK to use Google Translate. For any sort of business or commercial documents, or anything that has to be precise, it’s best to stick with human translators.
Here’s a head-to-head test from Verbalink that provides some findings and best practices of Machine Translation versus Human Translation.
Consider why machine learning and predictive analytics can provide top- and- bottom- line value to organizations like yours with the right tools, training, and processes for a range of objectives and use cases.