The listening lab is an open-source platform for audio annotation of less vocal and typically more challenging species.
This tool allows you to:
- Upload raw field recordings
- Automatic segment sparse features of interest to reduce processing time
- Train state-of-the-art transformer-based model for your application
- Export annotations and models for use in the field
- Outlier detection
This tool was developed by the University of Canterbury's Listening Lab Bioacoustic Research group https://github.com/listening-lab
Data available here Invasive Species dataset
Start frontend using cd frontend
then npm start
after installing dependences
Start backend server using cd backend
then uvicorn main:app --reload
Python dependences in ./backend/environment.yml
To start a clean build on your local machine run docker-compose up -d
after cloning the repository.
Install the latest images using docker hub docker image pull -a benmcewen/listening-lab
Start the frontend using docker run -p 3000:3000 benmcewen/listening-lab:frontend
and the backend using docker run -p 8000:8000 benmcewen/listening-lab:backend
We current use a transformer-based classification model - Audio Spectrogram Transformer (AST). The implementation can be found in ./backend/preprocessing/classifier.py
.
If you find this tool useful, please cite it
@article{mcewen2024active,
title={Active few-shot learning for rare bioacoustic feature annotation},
author={McEwen, Ben and Soltero, Kaspar and Gutschmidt, Stefanie and Bainbridge-Smith, Andrew and Atlas, James and Green, Richard},
journal={Ecological Informatics},
volume={82},
pages={102734},
year={2024},
publisher={Elsevier}
}
Feel free to contribute to this project or adapt it for your application.
- Prototype removed when label removed and unknown prototype not included at inference