New generative media models and tools, built with and for creators
The text introduces Veo, a highly capable model for generating high-definition video, and Imagen 3, a high-quality text-to-image model. These models have been developed in collaboration with the creative community to enhance the creative process.
Veo generates high-quality 1080p resolution videos in various cinematic and visual styles. It understands natural language and visual semantics, accurately capturing a prompt's tone and rendering details in longer prompts. Veo is available to select creators in private preview in VideoFX, and will be integrated into YouTube Shorts and other products in the future.
Imagen 3 generates photorealistic, lifelike images with fewer visual artifacts than previous models. It has an advanced understanding of natural language and can incorporate small details from longer prompts. Imagen 3 is available to select creators in private preview in ImageFX and will be coming soon to Vertex AI.
The text also discusses collaborations with the music community, including Grammy-winning musician Wyclef Jean, electronic musician Marc Rebillet, and Grammy-nominated songwriter Justin Tranter. These collaborations are informing the development of generative music technologies, including Lyria, an advanced model for AI music generation, and a suite of music AI tools called Music AI Sandbox.
The company is mindful of the challenges raised by generative technologies and is taking measures to ensure responsible deployment. This includes conducting safety tests, applying filters, setting guardrails, and developing tools like SynthID, which embeds digital watermarks into AI-generated content. All videos generated by Veo on VideoFX will be watermarked by SynthID.
