- Apply machine learning classifiers
- Stratified selection of ML detections for review
- Efficient keyboard-based annotation
- Grid-view and Focus-view annotation layouts
- Coming soon: Custom classifier training
Installation
macOS (Apple Silicon): open the .dmg and drag Dipper into Applications. Intel Macs are not supported.
Windows: double-click the installer and follow its steps. If Windows warns about an unrecognized app, click More info → Run anyway.
Linux, Ubuntu / Debian / Mint / Pop!_OS: download the .deb, then
sudo apt install ./Dipper_*.deb
Linux, Fedora / RHEL / openSUSE: download the .rpm, then
sudo dnf install ./Dipper-*.rpm (openSUSE: sudo zypper install ./Dipper-*.rpm)
Linux, other distributions or no admin rights: download the AppImage, run
chmod +x Dipper_*.AppImage, then run it. If it reports that FUSE is missing (Ubuntu 24.04 and
later), install it with sudo apt install libfuse2t64. The .deb and .rpm are the more reliable
choice where they apply: they use your system's own libraries and add Dipper to your app menu.
First use: to run models, install the machine learning environment from the Settings tab (a one-time 2.5–9.5 GB download). See the README.
Server Mode + Remote Access
To run Dipper on a server and access via a remote browser, see Server Mode instructions
ML Model Support
The app includes all major foundation models for bioacoustics, including BirdNET, Perch V2, and HawkEars
Screenshots
Video Tutorials
Download demo datasets here
About Dipper
Dipper is an open-source desktop application for bioacoustics machine learning workflows.
Annotation
Binary, multi-class, and bounding-box annotation of audio clips with spectrogram visualization.
Inference
Run trained models over large audio datasets and review detections in the built-in reviewer.
Bugs and Feature Requests
Report issues or request new features on our GitHub Issues page.
All Releases
Click any asset to download directly from GitHub Releases.
Loading releases…
