- Google’s video generation model obtained an important update
- Announced in Google I/O, I see 3 can combine audio and video at your output
- It is an ultra and only of the US function for now
Ia and Pika video generation tools can create alarmingly realistic video pieces, and with enough effort, you can join those clips to create a short film. However, one thing they cannot do is simultaneously audio. The new I see Google model can, and that could change the game.
Announced Tuesday in Google I/O 2025, I see 3 is the third generation of the powerful model of generation of Gemini videos. With the right notice, you can produce videos that include sound effects, background noises and, yes, dialogue.
Google briefly demonstrated this capacity for the video model. The clip was an CGI tier animation of some animals that spoke in a forest. The sound and video were in perfect synchronization.
If the demonstration can become use of the real world, this represents a notable turning point in the AI content generation space.
“We are emerging from the silent era of videos generation,” said Google Deepmind CEO, Demis Hassabis, in a press call.
Lights, camera, audio
It is not wrong. Until now, no other AI videos generation model can simultaneously deliver synchronized audio, or audio of any kind, to accompany the video output.
It is not yet clear if I see 3, which, as its predecessor, I see 2, should be able to generate 4K video, exceeds the current leader of Openi Sora videos in the video quality department. Google, in the past, has affirmed that I see 2 is an expert in producing realistic and consistent movement.
Anyway, the output of what seems to be fully produced video clips (video and Audio) can instantly see a more attractive platform.
It’s not just that I see 3 can handle dialogue. In the world of cinema and television, background noises and sound effects are often the work of Foley’s artists. Now, imagine that all you need to do is describe to see the sounds you want and attach to action, and generate everything, including video and dialogue. This is a job that leads the animators weeks or months.
In a release in the new model, Google suggests that he tells the AI ”a short story in his warning, and the model returns a clip that gives it life.”
If I see 3 you can follow the indications and the departure minutes or, ultimately, video and audio hours consistent, it will not spend much before we see the first animated function generated completely through I see.
I see is live today and is available in the USA. As part of the new ultra -level ($ 249.99 per month) in the Gemini application and also as part of the new flow tool.
Google also announced some updates to its videot generation model I see 2, including the ability to generate video based on reference objects that it provides, camera controls, surpassing to convert the portrait to the landscape and the addition of objects and erase.
@Techradar ♬ Original Sound – Techradar
You may also like
