AI Raspberry Pi cat detection and notification: get a text/email when your cat does something it's not supposed to do, and have AI narrate what it sees 👁️
This RPi starter kit makes it easy to get started with AI (both running device model & calling an LLM), and it's generalizable across other vision-related use cases.
Demo (sound on 🔊):
https://www.youtube.com/watch?v=8KSAiwyDoy8
Have questions? Join AI Stack devs
This starter kit is super simple: it allows you to use your Raspberry Pi to monitor what your cat does at home, and email you when your cat does something it's not supposed to do. My #1 use case is detecting if my cats are jumping on my dinning table or kitchen counter. You could also opt to turn this into an AI narrator, where the voice narrates on everything it sees.
Of course this doesn't just work for cats 😄. You are welcomed to simply change the prompts and do other kinds of detections:
- 🐦 for bird watching: email a summary of what birds came by during the day, for the birdies
- 🐻 Racoon deterrent (I have not tried this, but you can?): play loud sound when AI sees Racoons going through your trash cans. You need an external speaker attached to your RPi if you want to do this.
- 🪴 Plant monitor: email you when your plants are dying from not being watered
- 📦 Package alert: email when there's a new package delivered at the door, even better, AI can tell you if it's a big one
- (submit your fun use cases!)
- 🧠 Multi-modal models: OpenAI
- 📫 Notification: Resend
- 📢 Narration (optional): ElevenLabs
- 💻 Orchestration and function calling: Langchain
- 📖 Docs: Mintlify
- 🐱 Cats
To try out this starter kit, you need a Raspberry Pi with a camera module, optionally a speaker module for narration!
I can send you one if you are one of the first few lucky people to try this kit :). Discord DM me.
For detailed setup instructions, visit our docs page here (powered by Mintlify)
You can find helpful tips on setting up dev environment here
For common questions and bugs, feel free to try out our AI Q&A bot here. You could simply type on the search bar or use cmd+k shortcut.
