Active Learning for Bioacoustics BioDCASE Challenge

I am pleased to announce that the Active Learning for Bioacoustics BioDCASE Data Challenge launches on April 1st.

If you work with bioacoustics data and are interested in how machine learning models can learn efficiently from small amounts of labelled data with limited annotation budgets, this challenge is for you.

Participants will use BaseAL, our evaluation framework, to develop and benchmark active learning strategies on real bioacoustic datasets spanning both terrestrial and marine environments.

• BioDCASE task page
• Try the BaseAL demo

Thanks to the rest of the organising team: Lukas Rauch, Marek Herde, Shiqi Zhang, Rupa Kurinchi-Vendhan, John Martinsson and Sara Beery.