A photo from open sources
A new device translates a person’s speech and gives it out with the same in a voice. In the near future, all people will become real polyglot capable of speaking any existing language of the world. Microsoft has announced work on software, which will analyze and recognize any speech, convert it to the desired language and pronounce the translation in your own voice, writes Gizmodo. You can see the presentation of technology here. The company believes that the new product will break many stereotypes and will make a real revolution by breaking down the language barriers between by people. At the presentation, the device translated the speech of the main scientific English-speaking Microsoft employee Rick Rashid Mandarin Chinese with a delay of only few seconds. However, Rashid himself noted that the technology is still far from perfect and the company’s employees have a lot Work on the prototype to bring it to mind. Presumably, the development of the final version of the device will take up to five years. “There is still a lot of work in this area, but the technology is extremely promising, “Rashid said. He added that for Microsoft It’s important that the new system is not only fast, but also more accurate than other speech recognition technologies. “I believe, that we can use it to make the world a better place, “- concluded specialist. To date, linguists distinguish from 2 to 7 thousand different languages. These include natural languages, artificial, gestural and reconstructed. These numbers are more than approximate since a single approach to convention differences between languages and highlighting dialects within the same the language of linguists does not exist.
Arkady AZAEV
PHOTO: research.microsoft.com
