Microsoft has announced that it has created first machine human-level Chinese to English translation. It is one of a kind Artificial Intelligence (AI)-powered machine system that can translate sentences of news articles from Chinese to English with the same quality and accuracy as humans. Researchers from both USA and Asia said that their system has achieved human parity on a commonly used test set of news stories named newstest2017 which was developed by a group of industry and academic partners and released at a research conference WMT17 last fall. Post the tests, the company says that the results are accurate and on par with human-level. The team said that they have hired external bilingual human evaluators, who compared the Microsoft’s results to two independently produced human reference translations. The company is calling it a major milestone in one of the most challenging natural language processing tasks. The team used the dual-learning method meaning that a sentence which is sent to translation from Chinese to English, was also translated it back from English to Chinese. This method made sure the translation accurate, and it allowed the system to refine and learn from its own mistakes. Xuedong Huang, a technical member in charge of Microsoft’s speech, natural language, and machine translation efforts said: It ...
Check for tech updates, mobile phone launches, specifications and features of mobile phones, top best apps and lot more.
Advertisements
Thursday, 15 March 2018
Microsoft achieves first human-level Chinese to English machine translation using AI
Microsoft has announced that it has created first machine human-level Chinese to English translation. It is one of a kind Artificial Intelligence (AI)-powered machine system that can translate sentences of news articles from Chinese to English with the same quality and accuracy as humans. Researchers from both USA and Asia said that their system has achieved human parity on a commonly used test set of news stories named newstest2017 which was developed by a group of industry and academic partners and released at a research conference WMT17 last fall. Post the tests, the company says that the results are accurate and on par with human-level. The team said that they have hired external bilingual human evaluators, who compared the Microsoft’s results to two independently produced human reference translations. The company is calling it a major milestone in one of the most challenging natural language processing tasks. The team used the dual-learning method meaning that a sentence which is sent to translation from Chinese to English, was also translated it back from English to Chinese. This method made sure the translation accurate, and it allowed the system to refine and learn from its own mistakes. Xuedong Huang, a technical member in charge of Microsoft’s speech, natural language, and machine translation efforts said: It ...
-
Here are the social profile links that I have worked on for Backlinks https://docs.google.com/presentation/d/1gAvVQYafqO_jWsSIz3N95Sy3q5KAx...
-
The Optimus L4 II E440 owners may also want to keep their device up-to-date. So we thought of giving you a tutorial of how to upgrade Opt...
-
Vivo launched the vivo V17 smartphone late last year in India with a Super AMOLED display, 32MP in-screen camera, and more. The successor t...