Fichier de travail (INPUT) : ./DUMP-TEXT/2-11.txt
Encodage utilisé (INPUT) : utf-8
Forme recherchée : translation|traduction|机器
_________________________________________________________________________________________________
Ligne n°1 : ... #TechCrunch » Feed TechCrunch » Comments Feed TechCrunch » Microsoft- Ligne n°2 : announces breakthrough in Chinese-to-English machine translation
Ligne n°3 : Comments Feed alternate alternate ...- Ligne n°5 : Microsoft announces breakthrough in Chinese-to-English machine translation
Ligne n°10 : ... A team of Microsoft researchers announced on Wednesday they’ve created- Ligne n°11 : the first machine translation system that’s capable of translating news
Ligne n°12 : articles from Chinese to English with the same accuracy as a person. ...
Ligne n°14 : ... around 2,000 sentences from various online newspapers, comparing the- Ligne n°15 : result to a person’s translation in the process – and even hiring
Ligne n°16 : outside bilingual language consultants to further verify the machine’s ...
Ligne n°22 : ... It’s surprising, then, how quickly the researchers were able to achieve- Ligne n°23 : this milestone – especially given that machine translation is a problem
Ligne n°24 : people have been trying to solve for decades. ...- Ligne n°29 : “Hitting human parity in a machine translation task is a dream that all
Ligne n°30 : of us have had,” said Xuedong Huang, a technical fellow in charge of ...
Ligne n°30 : ... of us have had,” said Xuedong Huang, a technical fellow in charge of- Ligne n°31 : Microsoft’s speech, natural language and machine translation efforts,
Ligne n°32 : in Microsoft’s blog post. “We just didn’t realize we’d be able to hit ...- Ligne n°43 : But asking for a machine translation of a web page or news article
Ligne n°44 : still often renders the same hard-to-understand mess of words that, at ...- Ligne n°54 : “Machine translation is much more complex than a pure pattern
Ligne n°55 : recognition task,” said Ming Zhou, assistant managing director of ...
Ligne n°64 : ... Deep neural networks, a method of training A.I. systems, allowed the- Ligne n°65 : researchers to create more fluent and natural-sounding translations
Ligne n°66 : that take into account broader context that the prior approaches, ...
Ligne n°66 : ... that take into account broader context that the prior approaches,- Ligne n°67 : called statistical machine translation.
Ligne n°73 : ... The researchers said they used methods including dual learning for- Ligne n°74 : fact-checking translations; deliberation networks, to repeat
Ligne n°75 : translations and refine them; and new techniques like joint training, ...
Ligne n°74 : ... fact-checking translations; deliberation networks, to repeat- Ligne n°75 : translations and refine them; and new techniques like joint training,
Ligne n°76 : to iteratively boost English-to-Chinese and Chinese-to-English ...
Ligne n°76 : ... to iteratively boost English-to-Chinese and Chinese-to-English- Ligne n°77 : translation systems; and agreement regularization, which can generate
Ligne n°78 : translations by reading sentences both left-to-right and right-to-left. ...
Ligne n°77 : ... translation systems; and agreement regularization, which can generate- Ligne n°78 : translations by reading sentences both left-to-right and right-to-left.
Ligne n°80 : ... Zhou said the techniques used to achieve the milestone won’t be limited- Ligne n°81 : to machine translations.
- Ligne n°83 : “This is an area where machine translation research can apply to the
Ligne n°84 : whole field of AI research,” Zhou said. ...- Ligne n°86 : In addition, it could enable more accurate and natural translations
Ligne n°87 : across other languages in the future. ...- Ligne n°94 : But you can play around with the new translation system here on
Ligne n°95 : Microsoft’s website: https://translator.microsoft.com/neural. (This is ...
Ligne n°99 : ... The system will show a sentence in Chinese (simplified), which is then- Ligne n°100 : translated two ways, with the more perfected translation on the right
Ligne n°101 : to demonstrate the improvements. ...- Ligne n°103 : Machine translation is something researchers at Google have been
Ligne n°104 : working on as well, including with its own machine learning technique ...
Ligne n°110 : ... Image, top: Xuedong Huang, technical fellow in charge of Microsoft’s- Ligne n°111 : speech, natural language and machine translation efforts. (Photo by
Ligne n°112 : Scott Eklund/Red Box Pictures) ...