The term “neural” has become a keyword in the field of artificial intelligence recently as the companies are realising the importance of the latest advancements in machine learning. 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.
Neural Machine Translation essentially allows Google’s translation tool to take entire sentences into consideration instead of just translating one word at a time. “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,” the company said in its blog post.
As each sentence is translated in a better manner, it results in articles that are “smoother and easier” to read, the company said. The use of Neural Machine Translation also means that over time, the system will learn and produce more natural translations, the company claims.
Out of the total 103 languages supported by the Google Translate as of now, Neural Machine Translation is being implemented for translations to and from eight languages currently. These languages include English, French, German, Spanish, Portuguese, Chinese, Japanese, Korean and Turkish. The company said that these languages represent native languages of about one-third of world’s population and cover more than 35 percent of all Google Translate queries.
However, the company said that it plans to eventually roll out Neural Machine Translation to all of the 103 languages supported by its translation tool.
“With this update, Google Translate is improving more in a single leap than we’ve seen in the last ten years combined,” the company said. Further, the company’s cloud public cloud service, Google Cloud Platform, has made the system behind Neural Machine Translation available for all businesses through its Google Cloud Translation API.