Birth of the first female artificial intelligence broadcaster in South Korea: she can work 24 hours

November 10th. On November 6, South Korea’s MBN TV station launched South Korea’s first artificial intelligence (AI) anchor, and smoothly broadcast the main news and several news of the day. 

It is reported that the AI ​​female anchor is based on the TV station’s host Jin Zhuxia and named it “Jin Zhuxia.” As soon as it appeared, “Jin Zhu Xia” aroused public attention. Many netizens sighed after reading the news, “It can be used as false as the real” and “Awesome”.

The picture shows the AI ​​female anchor and the real male reporter

According to a report by South Korea’s “Joongang Ilbo” on the 10th, MBN TV station said that using AI anchors to broadcast news can quickly broadcast news to viewers in emergency situations such as disasters, and can continue to work 24 hours a day. It can save a lot of manpower, time and cost, and use it to try to produce new programs, which can effectively save resources.

Korean AI female anchor

How was “Jin Zhuxia” born? According to the report, “she” was a product born after a 10-hour recording of the “Comprehensive News” video actually hosted by Jin Zhuxia, and a long-term deep learning of Jin Zhuxia’s actions, sounds, and broadcast process. Quickly generate a 1000-word text broadcast video in one minute.

Kim Joo-ha and AI anchor

After the reporter made a broadcast draft of the news that occurred on the day, and edited the subtitles and video of the program director, the news screen was uploaded. Based on the data of deep learning, “Jin Zhu Xia” could imitate Jin Zhu Xia’s actual tone, tone and tone. The news is broadcasted in mouth.

Jin Zhuxia

It is understood that the “Jin Zhu Xia” AI anchor is jointly developed by MBN TV and the artificial intelligence development company “MoneyBrain”. The artificial intelligence video synthesis technology developed by the company integrates artificial intelligence, deep learning, and convolutional neural network (CNN) learning technology, which can vividly restore the way real people are speaking, which is indistinguishable. 

YCPress: