Extracting extended vocal units from two neighborhoods in the embedding plane
C. Lorenz, X. Hao, T. Tomka, L. Ruettimann, R. Hahnloser
The paper introduces a semi-supervised method to segment and cluster vocalizations by identifying two density-based regions in low-dimensional embeddings, achieving lower false positive error rates.
bioRxiv · Sep 26, 2022 · PreprintAnimal Behavior & Cognition