Foundation models for bioacoustics – A comparative review

Automated bioacoustic analysis is essential for biodiversity monitoring and conservation, requiring advanced deep learning models that can adapt to diverse bioacoustic tasks. This article presents a comprehensive review of large-scale pretrained bioacoustic foundation models and systematically investigates their transferability across multiple bioacoustic classification tasks. We provide an overview of bioacoustic representation learning by analysing pretraining data sources and benchmarks. On this basis, we review bioacoustic foundation models, dissecting their training data, preprocessing, augmentations, architecture, and training paradigms. Additionally, we conduct an extensive empirical study of selected models on the BEANS and BirdSet benchmarks, evaluating generalisability under linear and attentive probing. Our experimental analysis reveals that Perch 2.0 achieves the highest BirdSet score (restricted evaluation) and the strongest linear probing result on BEANS, building on diverse multi-taxa supervised pretraining; that BirdMAE is the best model among probing-based strategies on BirdSet and second on BEANS after BEATs N L M , the encoder of NatureLM-audio; that attentive probing is beneficial for extracting the full performance of transformer-based models; and that general-purpose audio models trained with self-supervised learning on AudioSet outperform many specialised bird sound models on BEANS when evaluated with attentive probing. These findings provide guidance for adapting models to new bioacoustic classification tasks.