whoBIRD is a mobile birding app that uses the BirdNET machine learning model to identify over 6,000 bird species worldwide from their songs and calls. All processing happens locally on the device, requiring no internet connection after the initial model download. It is designed for field use by birdwatchers, allowing real-time recognition anywhere from forests to remote lakes.
In the Computer vision, OCR & document AI space, whoBIRD takes a focused approach. It focuses on identifying unknown birds by their vocalizations in the field without needing an internet connection or manual lookup. It is built as an open-source project for birdwatchers and nature enthusiasts. The project is open source (GPL-3.0). whoBIRD is available on the web and Android.
It is developed by woheller69, and it first shipped in 2024. Development happens publicly on GitHub with 881 stars and 5 commits in the last 90 days. Key capabilities include bird sound recognition, offline identification, and over 6000 species. The interface is available in 26 languages, including Arabic, Tibetan, and Catalan.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do