Virtual broadcasters run on the world's first artificial intelligence, just like real people

Recently, China's news agency Xinhua has released two virtual television broadcasters (one speaking English, one speaking Chinese) operating on electricity, causing the world's tech community to be surprised.

Recently, China's news agency Xinhua has released two virtual television broadcasters (one speaking English, one speaking Chinese) operating on electricity, causing the world's tech community to be surprised. This announcer is a human digital image, using fully artificial voice to read the news and can work 24 hours without any communication platform.

Currently there is no information about the technology that Chinese engineers have used to create this newscaster, but many people believe that it may be the most advanced elements of machine learning research in the country. this people.

Picture 1 of Virtual broadcasters run on the world's first artificial intelligence, just like real people

It seems that the engineers have created a digital model from the image of a real announcer, then they use computer techniques to illustrate the aperture and facial expressions. It can be said that this announcer is a real electric puppet with the body being elements 1 and 0.

Xinhua can program virtual broadcasters to read news based on voice synthesized by computers. This method is faster than using traditional computer techniques.

The virtual announcer is the product of a joint project between Xinhua and Sogou to increase work efficiency and reduce production costs.

However, this radio broadcast is still limited. After watching the video, we can easily recognize that the announcer's voice is created with technology and the face has not shown much emotion.

But with more learning machine, this virtual announcer will soon be able to improve and reach the 'real' level.

See more:

  1. Microsoft artificial intelligence application can admire the poetic landscape, invite you to enjoy
  2. This is how AI defeats the best Dota 2 players
  3. Experience training AI with Ninja game, do you want to try?
Update 24 May 2019
Category

System

Mac OS X

Hardware

Game

Tech info

Technology

Science

Life

Application

Electric

Program

Mobile