Google's response to Sora from OpenAI did not take long to arrive - the company introduced the AI generator Veo, capable of creating high-quality videos.
Google claims that Veo has a "perfect understanding of natural language and visual semantics", which allows it to create any 1080p videos that can last "more than a minute." The generator also understands cinematic and visual techniques, such as the concept of framing.
Google showcased an example of collaboration with Veo - a promo created by the creative studio of American actor and comedian Donald Glover:
According to Google, Veo can simulate real physics better than previous models with significantly higher quality results.
"Everyone will become a director, and everyone must become a director," says Glover in the video, praising Google's advance. "At the heart of it all is a simple story."
From today, Veo will be available in the Google VideoFX tool for some creators, and the company says it will also appear in YouTube Shorts and other products. It sounds interesting, as the built-in video generator can outperform YouTube in the eternal battle with TikTok.
You can see more examples of Veo's work here.
Google also introduced Imagen 3 as its "highest quality" model for converting text to image - with an "incredible level of detail" for "photorealistic images" and fewer artifacts; and Music AI Sandbox - a set of tools that can help create songs and rhythms.
Comments (0)
There are no comments for now