paper
Beyond task performance: Decoding bioacoustic embeddings with speech features
paperactiveprovisional
beyond-task-performance-decoding-bioacoustic-embeddings-with-speech-features-babd7de1·1 events·first seen 2d agoAliases: Beyond task performance: Decoding bioacoustic embeddings with speech features
Co-occurring entities
More like this (12)
Acoustic Cue Alignment in Audio Language Models for Speech Emotion RecognitionSpeaker Group Encoding in Self-supervised Speech Recognition ModelsLeveraging Audio-LLMs to Filter Speech-to-Speech Training DataMulti-Faceted Interactivity Alignment in Full-Duplex Speech ModelsCross-Modal Masking for Robust Silent Speech Synthesis Using sEMG and Lipreadingbioacousticsbioacoustic monitoringListening with Attention: Entropy-Guided Explainability for Transformer-Based Audio Modelsspeaker-attribute classificationSemantic-Acoustic EquilibriumEthical and Technical Limits of Deepfake Speech DatasetsMassive Text Embedding Benchmark
Recent events (1)
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.