Google’s Neural Machine Translation System Makes Translations More Accurate

Google Translate is a very handy tool when you want to look up a word or get the gist of a phrase/sentence. However more often than not, thanks to the various subtleties in languages, the results of a translation might leave users more confused than enlightened, but Google thinks that they have a solution in the form of a Neural Machine Translation system.

According to Google, what makes the NMT system better is that it considers the entire sentence and the words and how they are related to each other, meaning that instead of translating each word without regard for context, it will consider that which in turn should hopefully make the end result a sentence that makes more sense, or one that seems less disjointed with seemingly random words.

Google explains, “Whereas Phrase-Based Machine Translation (PBMT) breaks an input sentence into words and phrases to be translated largely independently, Neural Machine Translation (NMT) considers the entire input sentence as a unit for translation.The advantage of this approach is that it requires fewer engineering design choices than previous Phrase-Based translation systems. When it first came out, NMT showed equivalent accuracy with existing Phrase-Based translation systems on modest-sized public benchmark data sets.”

That being said, machine-based translations are still far from perfect, but it looks like Google is hard at work at trying to improve them. Google expects that they will be working to apply its NMT system to more language pairings in the future over the course of the next few months.

Tyler Lee
Categories: General
Tags: Google