On May Tuesday, 2024, Google unveiled a new system called Veo at the Google I/O conference that can generate videos from text descriptions, similar to OpenAI’s Sora.
Veo can create high-definition videos that can last over a minute long and allow you to edit them with written prompts.
Both Veo and Sora are still under development, but they represent a new generation of AI tools that can create videos without needing to film anything. These video generators are yet to be developed, but they are becoming more capable all the time.
Veo appears to have similar features of Sora, but Google still needs to release Veo for the general users. Like with Sora, we can only base our impressions on the videos that Google has showcased.
As per Google, Veo can edit your existing videos. This can be an additional helpful feature where you can include other images to your videos.
It should help maintain visual consistency across the frames and create sequences that will last beyond a minute from a single text instruction or from a series of prompts given.
So, it is said to generate scenes and apply cinematic effects.
While Veo’s capabilities seem impressive, Google acknowledges that creating videos with AI is still challenging.
One drawback is that the objects and people in the videos can sometimes flicker or jump between frames. Google says it is working on improving Veo in this area.
Google plans to make Veo available to a small group of creators in the coming weeks. Eventually, Google hopes to integrate some of Veo’s capabilities into products like YouTube Shorts.
Also Read: Google’s Gemini AI Is Getting Third-Party Music Streaming Support