
- Category
- AI
- Website
- github.com
- Language
- English
- Listed
- 2d ago
- Stars
- 2,929
- Forks
- 73
arnegiacomo/fugleramme ↗, Python, MIT, Last commit 1d ago
A microphone feeds BirdNET-Go, which runs the BirdNET classifier locally, and Fugleramme turns its detections into a collage of the birds recently heard. Each species is matched to a cut-out from historic public-domain natural-history plates, over 800 of them across more than 400 species, hand-curated with none of the art AI-generated, and packed onto a textured paper page with larger birds toward the centre, sized by real body mass. The reference build is a Raspberry Pi 5 with a 13.3-inch Inky Impression panel in an A4 frame, though the e-ink panel is optional and the same view is served as a web kiosk. Because the plates are Scandinavian, British and central European, coverage is best for the Nordics, the British Isles and Germany. A live instance runs from a kitchen window in Bergen, Norway, showing the birds currently heard in that garden.