Based on artificial intelligence, WIPO Translates uses neural machine tranlsation technology, mirrors common language usage.
WIPO has initially “trained” the new technology to translate Chinese, Japanese and Korean patent documents into English. Patent applications in those languages accounted for some 55% of worldwide filings in 2014.
It offers an unprecedented level of accuracy: it is the result of the training of the neural machine translation tool that compared 60 million sentences from Chinese patent documents provided to WIPO’s PATENTSCOPE database by the State Intellectual Property Office of the People’s Republic of China with their translations as filed at the United States Patent and Trademark Office.
French is in the pipeline, other languages will follow.
The neural engine is integrated and outperforms the statistical engine.
Neural machine translation is an emerging technology. It is based on huge neural network models that “learn” from previously translated sentences. The specificity of neural machine translation (compared to previous “phrase based” statistical methods) is that it produces more natural word order, with particular improvements seen in so-called distant language pairs, like Japanese-English or Chinese-English.
Is the tool available to the public? Yes, the public can use the Chinese to English Engine on the Beta Platform.
To read the complete article, please click here.