{"id":6205,"date":"2018-03-14T06:11:06","date_gmt":"2018-03-14T13:11:06","guid":{"rendered":"https:\/\/blogs.msdn.microsoft.com\/translation\/?p=6205"},"modified":"2018-03-14T06:11:06","modified_gmt":"2018-03-14T13:11:06","slug":"human-parity-for-chinese-to-english-translations","status":"publish","type":"post","link":"https://www.microsoft.com\/en-us\/translator/blog\/2018\/03\/14\/human-parity-for-chinese-to-english-translations\/","title":{"rendered":"Neural Machine Translation reaches historic milestone: human parity for Chinese to English translations"},"content":{"rendered":"
<\/a><\/p>\n To validate th<\/span>e results, t<\/span>he researche<\/span>rs used a<\/span>n industry standard <\/span>test set of news stories<\/span> (<\/span>newstest2017<\/span>)<\/span> to <\/span>compare <\/span>human and machine translation<\/span> results<\/span>.<\/span>\u00a0<\/span><\/p>\n To <\/span>further <\/span>ensure accuracy of the <\/span>evaluation<\/span>, the team <\/span>also <\/span>hired bilingual human evaluators who compared the results against a different set of human-produced translations.\u00a0<\/span>\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0<\/span>\u00a0<\/span><\/p>\n R<\/span>ead more about this t<\/span>ranslation breakthrough<\/span> on <\/span>the <\/span>Microsoft Research blog<\/span><\/a>,\u00a0<\/span>and test (a portion) of this new system at <\/span><\/span>translate.ai<\/span><\/span><\/a>.\u00a0<\/span><\/span>\u00a0<\/span><\/span><\/p>\n For more in-depth <\/span>information about <\/span>the<\/span> neural machine<\/span> translation<\/span> technologies developed for this breakthrough<\/span>, <\/span>read the newly published research <\/span>paper<\/span><\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":" Microsoft announced today that its researchers have developed an AI machine translation system that can translate with the same accuracy as a human from Chinese to English.\u00a0 To validate the results, the researchers used an industry standard test set of news stories (newstest2017) to compare human and machine translation results.\u00a0 To further ensure accuracy of the evaluation, the team also….<\/span><\/p>\n
\nMicrosoft <\/span>announced <\/span>today<\/span><\/a> that its researchers have <\/span>developed a<\/span>n<\/span> AI machine translation <\/span>system<\/span> that<\/span> can translate with the same accuracy as a human from Chinese to English.<\/span>\u00a0<\/span><\/p>\n