Introduce the first 3D AI anchor
“她”来了!全球首位3D版AI组成主播精彩露脸
www.ef8ccz.com
BY wangxingwei from 21st Century
Published 2020-06-03
Xin Xiaowei is the world’s first 3D AI anchor. XINHUA
导读:她能随时改换发型,她能随时替换服装,她能络绎于演播室的不同虚拟场景中……但是,她却不是真人,她是全球首位人工智能驱动的3D版AI组成主播。

Xinhua News Agency introduced its first digital anchor in 2018. Developed by Xinhua and the search engine Sogou, this technology was developed through machine learning to simulate the voices, facial movements and gestures of real broadcasters. And now, Xinhua is expanding its lineup of virtual presenters. In May, Xinhua released the world’s first 3D artificial intelligence (AI) anchor, Xin Xiaowei.
新华社于2018年推出了首位虚拟主播。由新华社和搜索引擎搜狗一起研制的这一技能可以经过机器学习,模仿真人主持人的声响、面部表情以及姿态。而现在,新华社的虚拟主播部队中又增添了新成员。本年5月,新华社推出了全球首位3D人工智能主播新小微。

According to Xinhua, the 3D anchor is modeled on Zhao Wanwei, a flesh-and-blood reporter for the news agency.
据新华社介绍,这位3D主播的原型是新华社真人记者赵琬微。

Her digital counterpart was brought to life by “multi-modal recognition and synthesis, facial recognition and animation and transfer learning”, according to Sogou.
搜狗表明,这位赵琬微的“数码兼顾”是经过“多模态辨认及生成、实时面部动作生成及驱动、搬迁学习”等技能“熔炉”炼就而成。

Unlike previous virtual anchors, this newest addition “communicates visual and audio information more naturally and smoothly in real time”, Xinhua reported.
据新华社报导,和从前的虚拟主播不同,这位最新成员“在实时播报视听内容时愈加天然流通”。

Just as Xin said on her release footage: “In the future, I will walk out of the studio and bring you a refreshing news broadcasting style ...”
正如新小微在露脸视频中所说的那样:“未来,我将走出演播室,带给您全新的新闻播报方法……”

During her first appearance at an animated broadcast studio, she also performed complex facial expressions, for example, nodding her head and blinking as she spoke.
在虚拟演播室初次露脸期间,她还展示了杂乱的面部表情,如说话时不时允许或许眨眼。

More interestingly, Xin also can change her clothes and hairstyles based on different scenarios.
更风趣的是,新小微还能依据不同的场景改换发型服饰。

Though the new anchor already looks quite real, the agency said that the AI news anchor would be improved and updated in her expressions and movements. That will certainly create a new and dynamic viewing experience for their audience.
虽然这位新主播看起来现已适当传神,但新华社表明,这位人工智能新闻主播还会晋级改善表情动作。这无疑会为观众们带来全新动态的观看体会。

The AI anchor recently reported on the two sessions, the annual plenary sessions of the National People’s Congress and the National Committee of the Chinese People’s Political Consultative Conference.
这位人工智能主播近期还对两会(即全国人民代表大会和中国人民政治协商会议)进行了报导。

(Translator & Editor: Wang Xingwei AND Luo Sitian)
https://www.ef8ccz.com/story/3535.html
辞海拾贝
Anchor  主持人
   
Simulate  模仿
Gestures  姿态
   
Lineup  部队
Counterpart  对应的人
   
Multi-modal  多模态
Synthesis  组成
   
Transfer  搬迁
Audio  声响的
   
Footage  片段
Blinking  眨眼
   
Scenarios  场景
Dynamic  动态的
   
Plenary sessions  全体会议