This is a submission for the Google AI Studio Multimodal Challenge
What I Built
It is a trivia game that generates roastful questions about an image of a person you uploaded. As you can see, we use big Elon because why not? But you can totally upload yourself, and Gemini will try to challenge your hairline or something.
Demo
https://instant-game-show-host-452781430778.us-west1.run.app
https://www.loom.com/share/2e82e74b827b4225803b9d95c43b0f18?sid=a9469c24-3dd7-45e0-b1ce-abbb266ef8b0
How I Used Google AI Studio
I suffered. I suffered, but I used the AI studio. I made Gemini write all the code. It kept ruining it. I kept providing docs and curses to steer it. It kept trying to make my life miserable. It was a painful experience. Only the image of Pepe helped me in this endeavour. Thank you, Pepe. And, jokes aside, AI Studio is a nice tool that is just a bit rough on the edges. You should give it a try if you haven't.
Multimodal Features
The modalities used are text generation, image understanding, and live api. Image understanding is used to understand the photo, text generation is used to create trivia questions, and Live Api is used to make the conversation voice-enabled.
Top comments (0)