.github/ISSUE_TEMPLATE | ||
.ideas | ||
analyzer@08f0315852 | ||
config | ||
daemon | ||
media | ||
www | ||
.env | ||
.gitignore | ||
.gitmodules | ||
INSTALL.md | ||
install.sh | ||
LICENSE | ||
README.md | ||
requirements.txt | ||
TODO |
BirdNET-stream
Realtime BirdNET powered soundscape analysis for bird song identification.
Introduction
BirdNET-stream records sound 24/7 on any Linux computer with a microphone and analyze it using BirdNET algorithm by @kahst.
Bird contacts are stored in a database and are made accessible in a webapp.
Features
- 24/7 recording and BirdNET-Analyzer analysis of sound
- Live audio streaming and live spectrogram visualization from web browser
- Bird contacts saved into a SQL database
- Web Interface for system monitoring, data analysis and visualization
Requirements
BirdNET-stream aims to be able to run on any 64-bit Linux computer. It has been tested on Fedora and Debian.
It should work on a Raspberry Pi (or other Single Board Computer) with a USB microphone or Sound Card (not tested yet).
Installation
On debian based system, you can install BirdNET-stream with the following command:
curl -sL https://raw.githubusercontent.com/UncleSamulus/BirdNET-stream/main/install.sh | bash
For finer control, or to adapt to your system, you can follow the instructions in the INSTALL.md file.
Usage
-
BirdNET-stream web application can be accessed on any web browser at https://birdnet.home, from your local network, or at any other hostname you set in nginx configuration, if your public IP is accessible from the internet.
-
See the species detected
Acknoledgments
- BirdNET on which this project relies
- BirdNET-Pi the great inspiration of this project
License
BirdNET-stream is licensed under the GNU General Public License v3.0, see ./LICENSE for more details.