Is AI Already Influencing Translation? Technology and Artificial Intelligence (AI) in particular, is unashamedly breaking the language barrier, transforming translation as we know it. Global industry giant Amazon joined in the battle to produce the best translation technology for its customers with the launch of its own Amazon Web Series (AWS).
The service provides translations for a number of supported languages in a bid to help businesses expand their reach. While the convenience of automated translation applications cannot be disputed, one cannot help but ask whether new translation technologies are, in fact, needed or whether it just diversifies where our translations originate from.
AI Already Influencing Translation: The AI revolution is here
AI translation performed by artificial neural networks is already very much a part of our everyday lives, albeit unknowingly. At the forefront of the silent artificial intelligence revolution is Microsoft who, similar to Google, has developed its own translator app that does not only translate regular text but images, speech and street signs as well. The biggest drawcard of Microsoft’s app is undoubtedly its ingenious ability to run completely off-line – a huge benefit to individuals operating in areas with limited internet connectivity. Even Facebook’s translation service has been completely overhauled. Where it once focused on translating single phrases (which often resulted in very dubious translations) it now, thanks to AI, analyzed the entire context of a sentence, rendering increasingly accurate results.
Translation opportunities increased by technology
The most frequently used languages in the global market include German, Chinese, Russian, Spanish, and Portuguese, providing translators with some of their biggest job prospects. While it is often common belief that technology sets out to replace human labor, it has been found to, instead, create new opportunities for translators. Regardless of immense headway being made with regards to AI-assisted translations, the software still relies on human supervision to ensure that the exact dialect, grammar, and general translations have been utilized when decoding both text as well as voice recordings.
Accuracy of algorithms improving
Despite their algorithms becoming increasingly accurate it is still difficult for automated translators to compete with human excellence. An experiment was conducted at Sejong Cyber University in Korea in 2017 during which a group of human translators went head-to-head with 3 translator machines. Despite the hype surrounding them, the machines failed to live up to the expectations of the crowd, making significantly more mistakes with more than 90% of the texts being deemed ‘grammatically awkward’.
It is hard to predict to exactly what extent AI will permeate modern-day translations. It is, however, safe to assume that technological advances will continue to improve and revolutionize the industry as we know it, assisting humans in rendering faultless, timeous translation services.