Microsoft has created a new tool called VASA-1 that can turn a picture of a person’s face into a video where they talk or sing. The company says the lips move perfectly with the sounds, making it seem like the person in the picture is alive.
Even famous artworks like “The Mona Lisa” can be made to talk in VASA-1, speaking in modern accents.
But Microsoft is cautious about the technology being misused to impersonate people. VASA-1 works by taking a still image of a face and matching it with any person’s voice, real or fictional. It’s so advanced that the face in the picture moves while the voice is playing.
Researchers call VASA-1 a way to create lifelike faces for virtual characters. They believe it could lead to more natural interactions with AI avatars.
However, there are concerns about fraud. People might be fooled into thinking they’re talking to someone they trust when it’s actually a fake. Security experts warn that as this technology gets better, it’s important to be cautious about trusting digital interactions.
Microsoft says VASA-1 isn’t meant to deceive people, and they’re working on ways to detect fake content. They want to make sure the technology is used responsibly and doesn’t harm anyone.
While VASA-1 still has some flaws, Microsoft is hopeful that with further development, it could create even more realistic videos. They acknowledge that AI technology is improving quickly.
Source: Samaa