Perch 2.0 transfers 'whale' to underwater tasks

TL;DR

Perch 2.0, a bioacoustics foundation model trained on diverse species, demonstrates strong few-shot transfer learning performance on underwater and marine mammal audio tasks, outperforming previous models.

Contribution

This work introduces Perch 2.0, a new bioacoustics foundation model with broad species coverage, and evaluates its effectiveness in underwater tasks through few-shot transfer learning.

Findings

Perch 2.0 embeddings outperform other models in marine mammal tasks.

High performance of Perch 2.0 in few-shot transfer learning.

Recommended for developing classifiers with limited labeled data.

Abstract

Perch 2.0 is a supervised bioacoustics foundation model pretrained on 14,597 species, including birds, mammals, amphibians, and insects, and has state-of-the-art performance on multiple benchmarks. Given that Perch 2.0 includes almost no marine mammal audio or classes in the training data, we evaluate Perch 2.0 performance on marine mammal and underwater audio tasks through few-shot transfer learning. We perform linear probing with the embeddings generated from this foundation model and compare performance to other pretrained bioacoustics models. In particular, we compare Perch 2.0 with previous multispecies whale, Perch 1.0, SurfPerch, AVES-bio, BirdAVES, and Birdnet V2.3 models, which have open-source tools for transfer-learning and agile modeling. We show that the embeddings from the Perch 2.0 model have consistently high performance for few-shot transfer learning, generally outperforming alternative embedding models on the majority of tasks, and thus is recommended when developing new linear classifiers for marine mammal classification with few labeled examples.