Competitors size for Sora. While OpenAi allows you to generate ultra -theistic videos with this AI, Google does not want to stay out. A few months after I see 2, the company took advantage of its annual developer conference to present their successor, I see 3.
And this new video generator is distinguished from your predecessor with an important novelty: you can create videos with sound, including dialogue. As Google shows with several examples, these contents are very realistic, with Origami who merger, an old sailor who talks about the strength of the ocean or even an owl that arises and exchanges some words with a tuition before flying.
Detention results
From the beginning of his conference, Google showed the impressive capabilities of his new tool, with an opening film generated by View 3. Start with a “cowgirl” that reaches a small city, which seems empty. “This city has not seen anything yet,” he said, before music and many animals are launched (flamenco pink, elephant, giraffe …) appear one after another.
“In all areas, I see 3 stands out, from the textual and visual application to real physics through precise lip synchronization. It is excellent in understanding; you can tell a short story in your application and the model gives life,” Google said in a blog article.
It is enough for the user to indicate what they want in their request to generate the video. In addition to VI 3, Google has presented a new image generation model, presented as “the most efficient to date”, image 4. You can create realistic images clearly in terms of details (water drops, animal skins, etc.).
The American giant went even further by revealing the flow, a tool to make films driven by AI. Designed for creativity, it allows you to “easily create extracts, scenes and film stories combining the most advanced Google Deepmind models: I see, image and gemini.” This, with applications in natural language.
With the flow, users can also control the angles and movements of the camera or even modify existing plans. As a video shows, it is possible to extend a sequence that shows a woman running and arriving to a street with houses that have been destroyed to see her reach a street in her town with a party that takes place.
But I see 3 as a flow is not accessible for free. Only available in the United States, they can only be used by subscribers from Google AI Pro formulas (21.99 euros per month) and Google AI Ultra (274.99 euros per month).
Source: BFM TV
