Microsoft VASA-1: Microsoft has introduced VASA-1. VASA-1 is an AI tool that can create videos of human faces directly from still images. This AI tool can even synchronize the real expressions, emotions etc. of humans when it receives an audio clip. Microsoft has shared many samples of VASA-1 on its official website, and people will be really surprised to see it.
Microsoft VASA-1 AI Video Generator
The full form of VASA-1 is Visual Affective Skills Audio-1. This is a top AI model of the company, which has been specially designed for human facial expressions. It can generate a wide range of feelings and emotions through facial dynamics. This includes facial muscles, lips, nose, tilt of the head and many other factors.
Let us show you some video samples generated from Microsoft VASA-1;
Introducing: VASA-1 by Microsoft Research.
TL;DR: single portrait photo + speech audio = hyper-realistic talking face video with precise lip-audio sync, lifelike facial behavior, and naturalistic head movements, generated in real time.
Tap to see all the videos. pic.twitter.com/pPC6qZOBW2
— Eduardo Borges (@duborges) April 18, 2024
In this video you can see how by combining just a picture of a man and his speech, this AI tool has generated such a video, seeing which it is almost impossible to say that this video is not real.
— Eduardo Borges (@duborges) April 18, 2024
It is very easy to make fake videos
Similarly, in this video also you can see that by adding only one picture of this woman and her audio clip, this AI tool of Microsoft has made a 59 second video. Watching this video it seems that this is a completely real video.
Super easy to generate the videos.
Would you like to try VASA-1? pic.twitter.com/ogFjjKvOjt
— Eduardo Borges (@duborges) April 18, 2024
In this video, it is shown how using this new AI tool of Microsoft, a video can be generated with the real emotions and feelings of a person with the help of just a picture and audio clip. It is written in the caption of this video that with the help of this tool it is very easy to make such fake videos. By watching this video, you can understand yourself how a real-looking fake video of a human being can be created with the help of Microsoft VASA-1 AI video generative tool.
VASA-1 can currently only generate video at a maximum resolution of 512×512 pixels at 40fps. The company says that this tool has been designed to generate videos as close to real life as possible.
Also read:
Free Fire MAX OB44 Update goes live, see list of all new features and rewards
