README.md

metadata

title: Gemini Live Audio Chat
emoji: ⚡
colorFrom: blue
colorTo: purple
sdk: docker
manifest: /manifest.json
pinned: false
short_description: Audio chat with Gemini + tool calling + search + code run

This project demonstrates an interactive audio and text chat interface powered by the Gemini AI model. Users can speak to the AI or type text to communicate with it.

The AI is capable of calling external tools, searching for information on Google, and running code snippets. Function calling by the AI is performed on the server side, and results are sent back to the client. This approach is crucial for functions requiring access to server-side data or functionalities.

A Space for client-side function calling can be found at https://huggingface.co/spaces/Nirav121/gemini-live-ffc

The server uses Docker for deployment, making it easy to deploy on any platform that supports Docker.