The goal of this project is to be able to classify large batches of ARU recordings using transfer learning. The Perch framework is an incredible took for bird vocalization classifiation and works very well due to the fact that the final training is done on the same recordings that you want to classify.
In a spin-off from BirdNet Analyzer, we reimplement many of the same features BirdNet Analyzer has to offer, with a stronger focus on the agile modeling workflow. See the documentation below to see how to get started.
See the documentation.