
Today we introduced Lyria 3, @GoogleDeepMind's latest music generation model, which you can use to turn your ideas into musical tracks through prompts or even photos. Here are 3 tips (and fun facts…
The Hook
“Today we introduced Lyria 3, @GoogleDeepMind's latest music generation model, which you can use to turn your ideas into musical tracks through prompts or even photos.”
Product Reveal
“Here are 3 tips (and fun facts…”
Feature Teaser
“(No spoken dialogue — Electronic music with hip-hop beats and 808s, transitioning to country rock and Latin pop. Sound design includes risers and sub-bass drops.)”
Call to Action
“(No spoken dialogue — Electronic music with a final impact sound and a subtle fade out.)”
Synthesizable in Remotion · key components of the dynamic-ai-creative-launch template.
Text characters or words appear rapidly, often with a slight rotation or scale, creating a 'typewriter' or 'burst' effect. Background gradients subtly shift color.
Use `interpolate` for character opacity and position, `spring` for subtle scale/rotation on appearance. Implement `LinearGradient` for background color transitions, animating `stops` or `colors`.
Images quickly appear, scale up slightly, and then either hold or transition to another image with a quick cut. Often accompanied by a background color change.
Utilize `Sequence` and `AbsoluteFill` for rapid scene changes. Apply `spring` animations for image `scale` and `opacity` on mount. Background color changes can be achieved with `interpolateColor`.
Multiple instances of the same word or related words appear in a cascading or overlapping fashion, often with varying opacities and positions, creating depth and emphasis.
Render multiple `Text` components with varying `opacity` and `transform` properties (e.g., `translateX`, `translateY`). Animate these properties using `interpolate` with different `inputRange` values for a staggered effect.
Purpose. Introduce the product's core capability: turning ideas into music.
Execution. Rapid text generation in a UI, showcasing prompt-to-music, followed by a quick reveal of diverse images that can inspire music.
Purpose. Demonstrate the breadth of musical styles and customization options.
Execution. Fast-cut montage of images and text representing various music genres, moods, and languages, all animated with energetic transitions.
Purpose. Highlight advanced features like dynamics, tempo, and vocal languages.
Execution. Layered text animations emphasizing 'Dynamics', 'Tempo', and a grid of languages, showcasing granular control over music generation.
Purpose. Conclude with the product's unique identifier and call to action.
Execution. A visual representation of 'watermarked music' followed by a clear display of the product name and a call to learn more.
Objective
Launch a new AI-powered design tool that generates unique visual assets from text prompts and image inputs.How the recipe adapts
The DynamicTextReveal primitive would be used for prompt input, generating design concepts. The ImageGridPop primitive would showcase the diverse visual assets created, replacing music-related images with design mockups and abstract art. LayeredTextCascade would highlight design principles like 'Harmony', 'Contrast', and 'Balance'. The color palette would shift to reflect the design tool's branding, while maintaining the energetic motion language.Scene durations (12 scenes, 91s)
Key Remotion components
To ensure that artificial general intelligence benefits all of humanity

Gopuff introduces Go, an AI shopping assistant built with SpaceXAI: say what you need and the order is placed.

Iliad (Troy) trailer made by Grok Imagine 1.5, which was just released