SDLGASQMAug 2, 2025

Foundation Models for Bioacoustics -- a Comparative Review

arXiv:2508.01277v18 citationsh-index: 3
Originality Synthesis-oriented
AI Analysis

It provides guidance for practitioners in biodiversity monitoring and conservation on selecting and adapting models for bioacoustic classification tasks, but it is incremental as it focuses on comparative review and evaluation.

This paper reviews large-scale pretrained bioacoustic foundation models, evaluating their transferability across classification tasks on BEANS and BirdSet benchmarks, finding that BirdMAE achieves the best performance on BirdSet and BEATs_NLM is slightly better on BEANS, with both requiring attentive probing for optimal results.

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 overview bioacoustic representation learning including major pretraining data sources and benchmarks. On this basis, we review bioacoustic foundation models by thoroughly analysing design decisions such as model architecture, pretraining scheme, and training paradigm. Additionally, we evaluate selected foundation models on classification tasks from the BEANS and BirdSet benchmarks, comparing the generalisability of learned representations under both linear and attentive probing strategies. Our comprehensive experimental analysis reveals that BirdMAE, trained on large-scale bird song data with a self-supervised objective, achieves the best performance on the BirdSet benchmark. On BEANS, BEATs$_{NLM}$, the extracted encoder of the NatureLM-audio large audio model, is slightly better. Both transformer-based models require attentive probing to extract the full performance of their representations. ConvNext$_{BS}$ and Perch models trained with supervision on large-scale bird song data remain competitive for passive acoustic monitoring classification tasks of BirdSet in linear probing settings. Training a new linear classifier has clear advantages over evaluating these models without further training. While on BEANS, the baseline model BEATs trained with self-supervision on AudioSet outperforms bird-specific models when evaluated with attentive probing. These findings provide valuable guidance for practitioners selecting appropriate models to adapt them to new bioacoustic classification tasks via probing.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes