Building an Offline Arduino UNO Q Cyberdeck That Identifies Birdsong and Draws Vintage Field Notes

The B.L.O.O.M. (Bridging Local Observations, Openly Mapped) project introduces a DIY cyberdeck built on the Arduino UNO Q, designed for naturalists to capture field observations while offline. Housed in a vintage makeup case, the device integrates a USB camera, microphone, and HDMI screen to facilitate field note-taking. It leverages Edge Impulse for local birdsong classification and runs a Gemma LLM to assist with species identification. Data is stored in a local SQLite database, with an optional synchronization feature that uses Netlify functions to push content to GitHub. A key feature is the integration of Cloudinary’s image generation API, which transforms field notes into vintage-style illustrations using the Nano Banana model. This project emphasizes data privacy and local AI processing, providing a unique, hackable tool for field research that operates entirely without an internet connection until the user chooses to sync.