Google has introduced a newly updated machine-learning based system for its translator service - Google Translate. Announced via an official post on its Research blog, the company has updated its Google Neural Machine Translation (GMNT) tool with the addition of 'Zero-Shot Translation'.

According to Google, GMNT allows it to provide 'an end-to-end learning framework that learns from millions of examples' and significantly improves translation quality. But it's also quite a challenge for developers to add translation support for more than 100 languages.

After the Zero-Shot Translation update, Google will be able to perform translations between languages without the presence of a similar pairing instance in the past.

'In Google's Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation, we address this challenge by extending our previous GNMT system, allowing for a single system to translate between multiple languages. Our proposed architecture requires no change in the base GNMT system, but instead uses an additional "token" at the beginning of the input sentence to specify the required target language to translate to. In addition to improving translation quality, our method also enables "Zero-Shot Translation" — translation between language pairs never seen explicitly by the system' reads the post.

Furthermore, the post says that while Google may display similar results and analyses going forward, it hopes that the findings resonate with not only machine translation researchers, but also 'linguists and others who are interested in how multiple languages can be processed by machines using a single system'.

The post ends with the company adding that the new multilingual Google Neural Machine Translation system is up and running for all Google Translate users. According to Google, 'multilingual systems are currently used to serve 10 of the recently launched 16 language pairs, resulting in improved quality and simplified production architecture.

Courtesy by Gadgetsnow.com