Publication
Intermediate-Task Transfer Learning for Bioacoustic Data
Hannes Kath; Thiago Gouvêa; Daniel Sonntag
In: KI 2025: Advances in Artificial Intelligence. German Conference on Artificial Intelligence (KI-2025), located at KI-2025, September 16-19, Potsdam, Germany, ISBN 978-3-032-02813-6, Springer Nature Switzerland, 9/2025.
Abstract
The accelerating loss of biodiversity underscores the urgent need for up-to-date and reliable quantitative data to support evidence-based biosphere management. Passive acoustic monitoring (PAM) has emerged as a key technology for biosphere management, scalable wildlife monitoring in particular. While PAM facilitates large-scale data collection, efficiently analyzing the vast amounts of recorded data remains a significant challenge. This work explores the use of transfer learning and systematically investigates state-of-the-art models, comparing fine-tuning these models with using frozen layer weights, and evaluating the application of intermediate-task transfer learning. In intermediate-task transfer learning, a model pre-trained on data less related to the target data is first re-trained on a larger dataset more closely related to the target, before being re-trained on the target data itself. Our results show that fine-tuning improves performance compared to using frozen model weights, and that intermediate-task transfer learning is only beneficial for models trained on data significantly different from PAM data. These findings pave the way for developing a real-world, efficient PAM data analysis tool.