Perch 2.0: The Bittern Lesson for Bioacoustics
This work provides a robust pre-trained model for bioacoustics researchers, enabling improved species classification and transfer learning across diverse taxa, though it is incremental as it builds on an existing model.
The authors tackled the problem of bioacoustic classification by expanding Perch to a multi-taxa dataset with self-distillation and a new training criterion, achieving state-of-the-art performance on benchmarks like BirdSet and BEANS and outperforming specialized marine models in transfer learning.
Perch is a performant pre-trained model for bioacoustics. It was trained in supervised fashion, providing both off-the-shelf classification scores for thousands of vocalizing species as well as strong embeddings for transfer learning. In this new release, Perch 2.0, we expand from training exclusively on avian species to a large multi-taxa dataset. The model is trained with self-distillation using a prototype-learning classifier as well as a new source-prediction training criterion. Perch 2.0 obtains state-of-the-art performance on the BirdSet and BEANS benchmarks. It also outperforms specialized marine models on marine transfer learning tasks, despite having almost no marine training data. We present hypotheses as to why fine-grained species classification is a particularly robust pre-training task for bioacoustics.