- Advertisement -
- Google’s video -oriented model received a major upgrade
- Announced on Google I/O, VEO 3 can combine audio and video in the output
- For the time being it is an ultra and only function
AI Video Generation Tools such as such as Sora And Pika Can make alarming realistic pieces of video, and with sufficient effort you can tie those clips together to make a short film. However, one thing that they cannot do is at the same time generating audio. The new VEO 3 -model from Google can, and that can be a game changer.
Announced on Tuesday Google I/O 2025” Veo 3 is the third generation of the powerful Gemini video generation model. With the right prompt, it can produce videos with sound effects, background noises and, yes, dialogue.
Google briefly demonstrated this option for the video model. The clip was a CGI quality animation of some animals that spoke in a forest. The sound and the video were in perfect synchronization.
If the demo can be converted into real-world use, this represents a remarkable turning point in the space for generating AI content.
“We stem from the silent era of video appeal,” said Demis Hassabis, CEO of Google DeepMind in a press conversation.
Lights, camera, audio
He’s not wrong. So far, no other AI video generation model can simultaneously synchronized audio or audio of any kind, to accompany the video output.
It is still not clear whether VEO 3, which, like its predecessor, VEO 2, could be able to be able to perform 4k Video, surpasses the current leader of the VideoMeratie Openi Sora in the video quality department. In the past, Google has claimed that VEO 2 is skilled in producing realistic and consistent movement.
Anyway, performing what seems to be fully produced video clips (video And Audio) can immediately make VEO a more attractive platform.
It is not only that VEO 3 can handle the dialogue. In the world of film and TV, background noises and sound effects are often the work of Foley artists. Imagine now that you only have to describe to describe the sounds that you want to adhere to and attach to the promotion, and it performs it all, including the video and dialogue. This is work that animators need weeks or months to do.
In a release about the new model, Google suggests that you tell the AI ”a short story in your prompt, and the model gives you a clip that it brings to life.”
If VEO 3 can follow instructions and output minutes or, ultimately, hours of consistent video and audio, it will not be long before we look at the first animated function that was completely generated via VEO.
VEO is live and available today in the US as part of the new Ultra Tier ($ 249.99 per month) in the Gemini app and also as part of the new power tool.
Google also announced a few updates for his Veo 2 Video -generation model, including the possibility to generate video based on reference objects you offer, surpass camera control, to convert from portrait to landscape, and add and delete object.
@Techradar
♬ Original sound – TechRadar
Maybe you like it too
- Advertisement -