Welcome to the Pipecat + Gemini workshop! This repo will help you get set up to start building with Pipecat and Gemini. First up, let's get your dev environment configured and run a quickstart project.
- Python 3.10+
- Linux, MacOS, or Windows Subsystem for Linux (WSL)
- Download this repo to get started:
git clone git@github.com:daily-co/gemini-pipecat-workshop.git
cd gemini-pipecat-workshop
- Set up a virtual environment:
python3 -m venv env
source env/bin/activate
- Navigate to the quickstart directory:
cd quickstart
- Install dependencies:
pip install -r requirements.txt
- Create a
.env
file with your Google API key:
cp env.example .env
You can obtain a Google Gemini API key: https://aistudio.google.com/app/apikey.
A free Daily API key is provided for you for the duration of this workshop. Sign up for your own Daily API key at: https://dashboard.daily.co.
- Run the example:
python gemini-bot.py
You'll see a URL (http://localhost:7860) in the console output. Open this URL in your browser to join the session.
Once your Pipecat environment is configured and you've run the quickstart, there are two example projects to reference as you build out your Gemini bot:
For client/server apps, this is your starting point.
This starter contains client and server examples:
- Client: 6 different options to run your client, including
- Server: your Gemini bot, which will look very similar to the quickstart
Refer to the simple-chatbot README for more details.
For phone applications, this is your starting point.
This 63E6 project includes a Gemini server bot that you can call via a Twilio websocket connection.
Refer to the twilio-chatbot README for more details.
- Pipecat docs: Contains guides, server docs, and client docs for the client SDKs
- Gemini docs
-
- Web: A web-based game using Gemini Live, where the AI player tries to guess words your describe
- Phone: A phone-based version of the same game, where two AI players join you in a word guessing game
-
AI Engineering World's Fair Voice helper
- This bot is featured on the AI Engineer World's Fair home page. See the client and server code that powers it.