Inspiration
I wanted to build a portable AI-powered recording device that could capture personal stories and preserve important memories using affordable hardware.
What it does
Memory Radio records audio through a built-in microphone and saves it to a microSD card. It is also designed to use AI to organize recordings into meaningful memories.
How we built it
We built Memory Radio using an ESP32-S3 microcontroller with a 2.8-inch touchscreen, microphone, speaker, and microSD card. The firmware was developed using Arduino C++.
Challenges we ran into
One of our biggest challenges was connecting the ESP32-S3 directly to Google's Gemini API. We also faced difficulties integrating audio recording, storage, and wireless communication.
Accomplishments that we're proud of
We successfully built a working hardware prototype that can record and save audio onto a microSD card. We also developed a touchscreen interface to control the device.
What we learned
We learned that the ESP32-S3 is a small, inexpensive, yet powerful microcontroller with huge potential for AI and embedded systems applications.
What's next for Memory Radio
Our next goal is to integrate a lightweight local AI model that can process recordings without relying on cloud services, making Memory Radio more private and accessible.
Built With
- ESP32-S3
- Arduino C++
- 2.8-inch Touchscreen
- Microphone and Speaker
- microSD Card
- Wi-Fi
- Google Gemini API (AI integration)
Built With
- api
- esp32
- gemini
