Translation machines are extremely competitive devices since they offer the possibility of producing free translations from foreign languages. However, poor understanding of the target language often means we cannot tell whether it is interpreted properly, producing wrong or poor results.
Let’s see how tools like Google Translate work.
“When it generates a translation, Google Translate searches models in hundreds of millions of documents to choose the best translation for you. By identifying models in documents that have already been translated by humans, Google Translate can make intelligent guesses on how accurate a translation presumably is. This model search process through large quantities of text is called “statistical machine translation”. Not all translations will be perfect as soon as they have been machine generated. The more documents translated by humans which Google Translate can analyse in a specific language, the better the quality of the generated translation will be. This is why the accuracy of a translation will sometimes vary from language to language” (taken from the official Google Translate website).
Machine translation cannot grasp the nuances when adapting a text from one language to another. Anyone using a machine translation program must understand the target foreign language so they can immediately grasp and correct all the possible blunders generated by the program…or the final outcome will be funny at best!
So to sum up, machine translation is potentially an effective tool if you know its limits and if you have an adequate understanding of how to use it properly.




