Google presents Veo, Imagen 3 and Music AI Sandbox: AI models to create video, images and music |  TECHNOLOGY

Google has taken advantage of the framework of its I/O 2024 developer event to announce its new models of IA generative. The company has introduced Veo for video creation, Image 3 for image generation, and Music AI Sandbox for music composition.

The company has presented a series of updates as part of its commitment to AI, including Gemini, its new model, and its integration into other services. In addition, Google has announced other news, as detailed below.

I see: videos from text

Google has launched its alternative to Sora from OpenAI: Veo. This model is capable of creating high-definition videos in 1080p, lasting more than a minute and in various visual styles. Additionally, you can understand film terms like “timelapse.”

To avoid confusion with real videos, Google ensures that videos created with Veo will have an invisible watermark to identify them.

Image 3: images from text

The company has developed Image 3, a text-to-image AI model. During the presentation, its level of detail and its ability to handle text were highlighted, achieving good results by including fragments of text in images, something that other tools often fail to do. Image 3 also adds an invisible watermark to images to prevent misuse.

Prompt: A view of a person’s hand while holding a small clay figure of a bird in one hand and sculpting it with a modeling tool in the other hand. You can see the sculptor’s scarf. His hands are covered in clay dust. a DSLR macro image that highlights texture and craftsmanship. (Photo: Image 3)

Music AI Sandbox – Music Creation

Google has also introduced a set of music artificial intelligence tools called Music AI Sandbox. This model allows users to create music from scratch using different instruments. At the same time, the company has the Lyria model, a model for generating songs.

By Editor

Leave a Reply