TechPrufer
Get started
DeepMind
DeepMind logoDeepMind · Feature trail

Perch Bioacoustics Analysis Model

1 post · Aug 2025 – Nov 2025

What this trail covers

The Perch model is updated with state-of-the-art off-the-shelf bird species predictions and improved adaptation to new environments, particularly underwater. It is trained on a wider range of animals and anthropogenic noise, using nearly twice as much data. The model can disentangle complex acoustic scenes over thousands or millions of hours of audio data and answer diverse ecological questions. Its vector search library is integrated into tools like Cornell's BirdNet Analyzer, and it enables agile modeling for rapid classifier creation.

All DeepMind trails
  1. Three ways Google scientists use AI to better understand natureThis post details updates to the Perch bioacoustics analysis model, including state-of-the-art bird species predictions and improved adaptation to new environments like underwater. It is trained on a larger dataset and can disentangle complex acoustic scenes. The model's vector search library is integrated into tools like Cornell's BirdNet Analyzer.
  2. How AI is helping advance the science of bioacoustics to save endangered speciesThis post details an update to the Perch bioacoustics analysis model, enhancing its state-of-the-art off-the-shelf bird species predictions and improving its adaptation to new environments, especially underwater. The model is now trained on nearly twice as much data, encompassing a wider range of animals and anthropogenic noise. It can disentangle complex acoustic scenes over extended periods and answer diverse ecological questions. The vector search library is integrated into tools like Cornell's BirdNet Analyzer, and the post highlights the 'agile modeling' approach for rapid classifier creation.