Machine Translations: Good or Bad?

Nowadays, machine translation has become more and more ‘popular’ and more and more software programmers are trying to find new ways of improving it.

Machine translations are cheap, fast and useful for getting a general idea about what text written in a foreign language means. However, “general idea” isn’t always exactly accurate; the machine translator literally translates (word for word) the text which often results in grammatically incorrect, if not completely incoherent, text.

So, are they really that bad? See it for yourself.

Below are a couple examples of texts translated back and forth using Babelfish, Bing,, Google, and TransPerfect, with the final English translation:

1.       Original text (pulled from Yahoo news):

“H. was born with vein of Galen malformation, a rare brain condition in which the brain is missing the blood vessels — capillaries — that would normally slow blood flow from high-pressure arteries to low-pressure veins. Without capillaries, blood rushes to those veins, causing them to expand and allowing blood to pool where it’s not needed.”

          …8 translations later, Bing gives us:

“In a vein of Galen malformation, H. – usually slows down blood flow to the brain, high blood pressure of the capillary venous pressure, was born with a rare exception. Spread of pools of blood and capillary blood is not necessary to get a vein.Capillary blood vessels to expand and make the blood Rush was required, which is caused by them.”

                    …2 translations later, TransPerfect gives us:

” H. Be born with the vein of Galls malformation, an uncommon brain condition in which the brain are missing the blood barques – – capillaries – – that wanted to flow ordinary slow blood from hogedruk arteries to low-insistence veins. Without the capillaries, blood insistences to that veins, provocation them to extend and whereas blood to pool where it will not needed. “

2.       Original text (pulled from “Twilight” book):

“I stared at the deep-voiced boy, taken aback, but he was looking away toward the dark forest behind us. He’d said that the Cullens didn’t come here, but his tone had implied something more—that they weren’t allowed; they were prohibited. His manner left a strange impression on me, and I tried to ignore it without success.”

          …6 translations later, SDL gives us:

“I considered I put the I express deep of boy, blocked, but itself devious the eyes after the dark forest behind those. It had said that the Cullens does did not come here, but its tone had not been permitted that something implied more he qu ‘; was prohibited. His way left a strange impression in me and in experienced negligenciá-he without the success. “

                    …8 translations later, Bing gives us:

“I am surprised that the kids deep voice, but he looked at the dark forest behind us. It is not here, but her voice, says Karen more, does it mean that it is not allowed. They are disabled. Let me, I tried a strange impression, style and ignore it. “

Not so good, right?

We believe that bad translations create a number of problems, such as damaging your brand and resulting in unhappy users. Moreover, there are many translation jobs where accuracy is absolutely critical, such as legal documents.  And translations of literature, poetry, and the like will remain difficult for machine translation software for years since there is much more to this sort of translation than accuracy, such as style and other artistic considerations.

On top of that, machine translation will never become as professional as a human translator can be. It will be always necessary to rely on human translators who offer specialized and professional if we want the output quality to be high.