Mastering language: Machine learning helps Google Translate become more accurate

Mastering language: Machine learning helps Google Translate become more accurate

The company has now incorporated a new Deep Learning version of Translate that will be rolling out in eight languages to start.

In early 2017, the Google Cloud Platform will offer GPUs worldwide for the Google Compute Engine and Cloud Machine Learning users.

Thankfully, translation services - such as Google Translate - removes barriers, allowing people to more easily communicate despite language differences.

In addition to the nine languages supported by the Microsoft Translator speech API, namely Arabic, Chinese Mandarin, English, French, German, Italian, Brazilian Portuguese, Russian and Spanish, neural networks also power Japanese text translations. The projects include a game that guesses what you're drawing, a camera app that recognizes objects you put in front of it and a music app that plays "duets" with you.

With the newly-rolled out update, Google Translate is said to have improved its translation service in a single stride than what we've seen over the last 10 years.

There are not many startups that are capable of competing with Google Translate, especially now with this upgrade.

Malaysia pins Pacific trade pact hopes on Abe-Trump meeting
The newspaper said Mr Trump was working without official State Department briefing materials in his dealings with foreign leaders. Abe may also try to sway Trump on the Trans-Pacific Partnership, a 12-country trade agreement that the president-elect opposes.

Those languages cover about one-third of all the translation requests that Google gets.

The API comes along with the announcement of a new Cloud Machine Learning group at Google, which will be led by machine learning experts Fei-Fei Li and Jia Li, the post said.

Neural Machine Translation has been producing exciting and wonderful research results in the past few years.

Google explains Neural Machine Translation as a system that "translates whole sentences at a time, rather than just piece by piece".

Of course this is just the first phase as their goal is to eventually roll out the Neural Machine Translation to all of the 103 languages and and surfaces that use Google Translate. Several apps have already made use of neural machine learning to improve and now Google has implemented "Neural Machine Translation" to make its translation tool better than ever before.

For example, if you're trying to draw an apple then the machine learning algorithms will read and analyse the path you followed to draw it.