Clone the repo and create python environment with the required dependencies.
python -m venv venv
source venv/bin/activate
python -m pip install --upgrade pip
pip install -r requirements.txt
# You might need ffmpeg/libsndfile or other system dependencies to get librosa working
Download & extract Wave
Optional Download all the required audio chunks and pretrained model from kaggle.
kaggle datasets download gaborfodor/rainforestexampleaudiochunks
mv rainforestexampleaudiochunks.zip data/
unzip data/rainforestexampleaudiochunks.zip
- Start the Wave development server:
$ ./waved - Start the app:
$ wave run rainforest.dashboard - Go to http://localhost:10101/
Select any species and fix the collected labels.
Upload your own rainforest recording and analyze it with a single pretrained CNN14 model.
