Google Translate now support 103 languages from just few supported at the time of launch 10 years ago, and was using statistical models to translate text—today, introduces even better “Neural Machine Translation,” which translates complete sentences, rather than just piece by piece.
Neural Machine Translation, at a high level, translates the whole sentences at a time, and by using this broader context, it figure out the most relevant translation, rearranges and adjusts and produce with proper grammar, more like a human speaking.
Since, it understand each sentence, the translated paragraphs and articles are a lot smoother and easier to read, google writes. And, because the end-to-end learning system is built on Neural Machine Translation, the system learns over time to create better, more natural translations.
The Neural Machine Translation, starts today with eight supported language to and from within Google Search, Google Translate app, and website—including: English and French, German, Spanish, Portuguese, Chinese, Japanese, Korean and Turkish.
Google said, these native languages represent around one-third of the world’s population, covering more than 35% of all Google Translate queries. However, they said, eventually Neural Machine Translation will be rolling out to all 103 languages and surfaces where ever Google Translate is accessible.
In addition, Google’s public cloud service Google Cloud Platform has today began to offer “Machine Learning APIs,” making it easy for anyone to use machine learning technology.
And, even more, businesses can also now access, the system behind Neural Machine Translation through Google Cloud Translation API.