Advances in neural machine translation system are behind new web and mobile versions of Google Translate that can now translate Chinese into English in entire sentences.
Google is intent on winning the war for leadership in the world of artificial intelligence, and one of the most coherent ways of doing this is by breaking the language barrier.
The Google Neural Machine Translation system (GNMT) take on a sentence as a whole and translates this into another language.
Evolution of Google Translate
The company said since it first launched Google Translate 10 years ago, rapid advances in machine intelligence have improved its speech recognition and image recognition capabilities.
Google said that GNMT achieves the largest improvements to date for machine translation quality.
In a new technical paper, Google research scientists Quoc V. Le and Mike Schuster from the Google Brain Team chart the move from phase-based machine translation to neural machine translation, which considers entire input sentences as a unit for translation.
“Using human-rated side-by-side comparison as a metric, the GNMT system produces translations that are vastly improved compared to the previous phrase-based production system,” the researchers said.
“GNMT reduces translation errors by more than 55pc-85pc on several major language pairs measured on sampled sentences from Wikipedia and news websites, with the help of bilingual human raters,” the researchers said.