Gemini Decisions API demo answers audio questions near real-time across all modalities
AI_Andrew · x · 2026-11-13
Developer thorwebdev demoed the Gemini Decisions API, which handles text, images, video, documents and audio — and can answer questions about audio in near real-time.
He built an animal-sounds guessing game with it: the app hears an audio clip and near-instantly identifies the animal, showcasing the API's real-time audio understanding. A live demo link is included.
More from Multimodal
- Fable 5.2 codes a pixel-art scene into an explorable interactive 3D world — dfeinition · 2026-11-13
- Claude builds 3D sets via MCP, Nano Banana and Seedance render an action teaser — azed_ai · 2026-11-13
- Liquid-metal cityscapes and warped portraits: a striking set of AI-generated images — gizakdag · 2026-11-13
- WIP AI-generated 3D bird model shows strikingly realistic feather detail — thespite · 2026-11-13
- MiniMax H3 video generation smears on motion for low-VRAM users, no fix found yet — CharacterSpinach3306 · 2026-11-13
- AI-Generated 'The Office' Skit Puts Jensen Huang to Work at Dunder Mifflin — AIandDesign · 2026-11-13