Almanac
paper

Beyond task performance: Decoding bioacoustic embeddings with speech features

paperactiveprovisionalbeyond-task-performance-decoding-bioacoustic-embeddings-with-speech-features-babd7de1·1 events·first seen 2d ago

Aliases: Beyond task performance: Decoding bioacoustic embeddings with speech features

Co-occurring entities

More like this (12)

Recent events (1)

3arXiv · cs.LG·2d ago·source ↗

Probing bioacoustic embeddings for speech-like acoustic features reveals no-free-lunch pattern

A new arXiv preprint investigates which acoustic features are encoded in pretrained bioacoustic audio embeddings using 88 eGeMAPS speech features across six taxonomic groups. Linear and nonlinear regression probes reveal that no single model captures the full acoustic feature space, with loudness best recovered (R²=0.76) and fundamental frequency hardest (R²=0.33). A concatenated embedding approach achieves highest overall performance, suggesting complementary coverage across models. The work provides data-driven model selection guidance for bioacoustics tasks involving rare species or low-resource domains.