Google Is Using Neural Nets To Translate Some Languages

0
Shares

Google Translate is one of those services rendered by the giant Internet company that come in handy when in a situation where you really want to understand what a person just wrote or spoke.

Language is still complicated for machines and the system is still not upto the level human beings are able to use it. Some of the translations end up seeming incoherent which is easily noticed by the regular user of the language.

That is why Google started to dip its toes on Neural Machine Translation on September last year which they said it “utilizes state of the art training techniques to achieve the largest improvements to date for machine translation quality.”

Google also explained how the Neural machine translation works:

At a high level, the Neural system translates whole sentences at a time, rather than just piece by piece It uses this broader context to help it figure out the most relevant translation which it then rearranges and adjusts to be more like a human speaking with proper grammar..

In November, Google said that they were putting the system into action with a total of 8 language pairs to and from from English and French, Spanish, German, Korean, Japanese, Chinese and Turkish. Now, they have added Hindi, Russian and Vietnamese.

The company says that you will find these new translations on the iOS and Android apps as well as Google search, Google app and on translate.google.com so expect better translations with this update. Google also promises to bring these neural machine translation to more languages over the next few weeks.

Shares