Let's go Bananas with GenMedia
summary
A public‑domain book goes through the full gen‑media stack live: Gemini as prompt engineer reads the text and writes prompts, Imagen portraits (uploader spelling) produce character images, Veo uses those images as first frames to animate video, Lyria composes a different piece per chapter and Lyria Realtime performs like a DJ‑style continuous music generator, and the TTS trick maps two base voices into four distinct characters. Gemini‑written prompts are part of training data, interactions API caches context to cut multi‑turn cost, and a demo run costs about $1 with video at ~5¢/sec.
AI-written summary · about
chapters (11)
- 0:00Introduction to role
- 2:44DeepMind media models
- 6:34Workshop project setup
- 10:52Technical configuration
- 15:34Chat and API context
- 21:26Character generation
- 26:10Illustrating book chapters
- 34:13Animating with Veo
- 39:51Music generation with Lyria
- 47:54Text-to-Speech narration
- 54:08Real-time demos and Q&A
related talks
27mBuild creative apps with the GenMedia suite
56mBuilding Multimodal Apps with Gemini 2 0 and Gemini SDK
22mGemini: Slashing Subtitle Time & Cost with AI
01h 00mHow to work with generative AI in JavaScript
21mBuilding at the Speed of Thought
18mIntroduction to Gemini APIs and AI Studio
44mHands-on-lab: It's all about going hands-on with Generative AI!
01h 45mAI Engineering with the Google Gemini 2.5 Model Family