A group of Microsoft researchers said that they use artificial intelligence (AI) in machine transcriptor to achieve equal accuracy in the field of Chinese and English translations.
Researchers from the Asia and US labs of the organization said that their translation system has achieved equal expertise in the examination of 'Newslett 2017'. In this development, the industry related professionals and academic partners participated in the development of the translation system. WMT 17 - The results of the translation system were unveiled at the research conference.
In order to verify the accuracy of this result, interpreter evaluators were appointed from outside, who compared the results of Microsoft to the results of comparative evaluation of results.
Researchers have been working for the last four decades to solve machine translation problems. Also, the researchers said that achieving this milestone proved that machine translation is a breakthrough solution in this case.
Ming Zhou, Assistant Managing Director of Microsoft Research Asia and head of the Natural Language Processing Group, working on the project, said that we are excited about the milestone of becoming the equivalent of people in the dataset. However, there are still many adversities in this regard, such as the results verification system in real-time news stories.
However, academic researchers and researchers have been working on translations for many years. However, they have recently achieved significant success using the training AI system entitled Deep Neural Networks. Which gave them an opportunity to spontaneously translate into a greater context than ever before. Known as the Statistical Machine Translation.
Microsoft's three research teams worked together in Redmond, Beijing and Washington to improve the translation system using training methods, as well as the fields to add to the milestone in achieving the milestone of achieving the equivalent of human equivalents in Datasets. In many cases, the method will follow the pronunciation of the person until it is absolutely correct.
Dual learning is the same method Microsoft uses in machine translation. It works like a verification of information. Each time the system gives a sentence for translation from Chinese language to English, the research team translates sentences again from English to Chinese. People usually use this method to verify the accuracy of translations. Besides, it will give the system a chance to make the wrong decisions. Microsoft's research team has also developed dual learning methods. This innovation can be used with artificial intelligence for other purposes.
Another method is, Deliverance Networks. It's like editing and correcting people's own writings repeatedly. Researchers have taught the machine how to repeat the same sentence in the field of translation until translation has been defunct.
Another method is the use of joint training techniques in English translation from English to Chinese and Chinese to English translation in machine translation. This method is used to translate new English sentences from English to Chinese language translations and to connect with the new sentence in Chinese language. This is the same process from English to Chinese. This will increase efficiency in two types of translations.
Another method used in machine translation is the agreement regulation. Through this method, the system can translate from left to right or right to the left. If this technique is used in the same translation, then it makes the translation results credible. This technique is very effective in maintaining proper rhythm of translations.