Post by Google DeepMind

Our foundational bioacoustics model Perch 2.0 developed with Google Research can be used to understand underwater marine ecosystems. 🌊 To protect the planet’s species, we need to locate and identify them at scale. Last year we launched Perch 2.0, trained primarily on bird sounds. Now, it’s successfully transferred bird data to underwater sounds because of its ability to: 🔄Generalize: it can expertly categorize sounds not included in its training. 🔊Classify similar sounds: it’s forced to learn detailed acoustic features. 👁️‍🗨️ Recognise diversity: identify a variety of species that have evolved vocalizations, yet still share structural characteristics In training, Perch 2.0 was evaluated on a range of whale vocalization tasks: distinguishing different baleen whale species, and killer whale subpopulations. Compared against pre-trained models, it landed consistently in the top performing for each dataset and sample size. To support this research further, we’ve created an end-to-end demo for working with the data, supporting new discoveries at scale. Find out more →  https://goo.gle/4trNGKW