Project Hub /IoT & Connected Devices/Telemetry & Remote Logging/Memento Cognitio: AI-Powered Visual Cognition Device

Intermediate

AI / ML

Project Specific

3 – 6 Hours

Partial Kit

Memento Cognitio: AI-Powered Visual Cognition Device

An AI-powered visual cognition device built with the Adafruit MEMENTO Camera Board that analyzes captured images using Google Gemini and speaks descriptions via Bluetooth.

Originally published on CircuitDigest

View original tutorial

Parts You'll Need

This project is sourced from another site, not written by us — this list is our best match of their components against what we stock, not a verified bill of materials. Always check the original blog (linked above) for the actual parts list before buying.

ESP32 CAM WiFi Module Bluetooth with OV2640 Camera Rs. 1,483.00
400 Points Mini Solderless Prototype PCB Breadboard High Quality Rs. 89.00
MB102 Adjustable Breadboard Power Supply Module (3.3V & 5V) for Arduino & Raspberry Pi Rs. 139.00
Jumper Wire 10 Pin Male To Male 20cm Rs. 45.00
Jumper Wire 10 Pin Male To Female 20cm Rs. 45.00

Selected: Rs. 0.00

We don't carry yet: Battery (generic / unspecified)

Browse all components

How It Works

The Memento Cognitio functions as an intelligent, portable visual assistant by integrating hardware image capture with cloud-based AI processing and local speech synthesis. At its core, the system relies on the Memento DIY camera main board, powered by the USE-803048 1100 mAh rechargeable battery, to capture photographs of the user's immediate environment. Once an image is recorded, Python 3.12 and CircuitPython manage the execution environment on the device, interfacing with external APIs to send the captured visual data to Google Gemini for real-time analysis and description generation. The resulting descriptive text is then processed using a Piper TTS model (specifically the en_US-libritts_r-medium.onnx file) alongside the pydub library to convert the textual response into spoken audio. Finally, the audio stream or control signals are transmitted to external output devices using the bleak library for Bluetooth communication, allowing the user to hear real-time audio descriptions of their surroundings.

Why Build This

Develop practical assistive technology applications that provide real-time visual assistance and descriptive feedback for visually impaired users.

Gain hands-on experience integrating embedded hardware like the Adafruit MEMENTO camera with advanced machine learning APIs and local text-to-speech models.

Master wireless peripheral communication and audio pipeline management using Python-based tools such as bleak and pydub.

Real-World Application

Providing accessible, real-time visual assistance and descriptions of surroundings for visually impaired users.

Skills You'll Learn

Bluetooth low energy communication

text-to-speech pipeline configuration

API integration for computer vision

battery-powered embedded systems management

audio processing with Python

Safety Precautions

No specific precautions noted for this build — always follow general electronics safety practices.

Technology Tags

CIRCUITPYTHON

GOOGLE GEMINI

BLUETOOTH LOW ENERGY

BLE

TEXT-TO-SPEECH

PIPER TTS

IMAGE RECOGNITION

OCR

Frequently Asked Questions

Do I need a continuous internet connection for the device to work?

Yes, while text-to-speech synthesis can run locally depending on your deployment, the image analysis step relies on querying the cloud-based Google Gemini API, requiring an active network connection.

How is audio output delivered to the user?

Audio descriptions generated by the Piper TTS model are transmitted wirelessly to compatible Bluetooth audio receivers or headphones using the bleak library.

Can I use a different battery instead of the USE-803048 1100 mAh model?

You can use an alternative rechargeable lithium polymer battery, provided it matches the voltage requirements and physical space constraints of the MEMENTO camera board.

Gallery

This page is part of Kitkraft's Project Hub — a curated collection of electronics & robotics projects gathered from the maker community to help students and hobbyists discover and learn to build. Kitkraft does not claim authorship of the original project, and makes no representation as to the accuracy, completeness or safety of third-party source material. The summary and "How It Works" explanation above are AI-generated, rewritten with reference to the original tutorial — not a reproduction of it. Full credit and a direct link to the original source are provided above. If you're the original creator and have a concern about how your work is represented here, please contact us and we'll address it promptly.

Kitkraft Project Hub — curated from the maker community, credited at the source.

PRJ-01348