Let’s face it; we have been hearing about the emergence of artificial intelligence for several decades now — with the promise of machines that have the intelligence to rival that of a human beings. Until recently, this has been predominately the focus of wishful thinking; however, recent developments have produced some promising results.
One area in which artificial intelligence is becoming increasingly prevalent, is in the automobile industry. There are cars that have the capacity to park themselves, with little to no assistance. Some cars can auto-correct accidental drifting into other lanes, or even stop when an object is detected in front of the vehicle. Additionally, Google has developed a vehicle that has the capacity to drive itself for thousands of miles without human assistance, without the occurrence of an accident.
More recently, Microsoft and Google released their version of artificial intelligence applications that have the capacity to translate both, audio and textual content from one language to another. These applications can be downloaded to a smartphone, tablet or desktop.
Interested to see how these applications measure up to standards set by a human translator? One of the top translation firms decided to put this technology to the test. The key was to determine if this type of program had the capacity to detect the subtle nuances that exist in language communication, which impact interpretation and translation. The firm determined that they would use Spanish as the language to be translated into English, so they selected two of their top Spanish translators to participate in the test. One translator, Gaby, would serve as the judge between the human translator, Adriana, and the Google’s translation tool.
This test focused on several key elements that are vital to producing an accurate translation from one language to another, including cultural idioms, grammar, and comprehension. Additionally, general strengths and weaknesses were given gravity in this assessment as well.
Surprisingly, the Google Translator was able to produce a translation of the audio and written content in a relatively quick manner. However, the program lacked the ability to efficaciously detect the slightest nuances in the content that allows for the most accurate translation. On the contrary, Adriana was able to not only detect those nuances, but she was able to apply them in order to produce a highly accurate translation of the content.
When it comes to man versus machine or technology, the human element is still superior when it comes to intellect. The Google Translator had several deficiencies that were immediately apparent, such as the fact that it has a proclivity to be hyper-literal when translating the text or audio content. This is due to the fact that Google does not have the capacity to detect certain cultural idioms or nuances that are indicative of how a particular word is used. This also leads to poor word choices in the translations. Transliteration rarely provides the most accurate translation.
Another area that this program struggled was in the area of grammar — not having the ability to translate the difference in grammatical presentation between the two languages. Additionally, there was also a problem with fractured syntax, which further dilutes the lucidity of the translation. Conversely, the human translator fared exceptionally well in the areas in which the artificial program was weak. At least for now, there is still no substitution for human intelligence, especially when it comes to language translation.