KOPS - The Institutional Repository of the University of Konstanz

A practical guide for generating unsupervised, spectrogram-based latent space representations of animal vocalizations

A practical guide for generating unsupervised, spectrogram-based latent space representations of animal vocalizations

Cite This

Files in this item

Checksum: MD5:616c93357904601a2957d34f743bf54b

THOMAS, Mara, Frants H. JENSEN, Baptiste AVERLY, Vlad DEMARTSEV, Marta B. MANSER, Tim SAINBURG, Marie A. ROCH, Ariana STRANDBURG-PESHKIN, 2022. A practical guide for generating unsupervised, spectrogram-based latent space representations of animal vocalizations. In: Journal of Animal Ecology. Wiley. 91(8), pp. 1567-1581. ISSN 0021-8790. eISSN 1365-2656. Available under: doi: 10.1111/1365-2656.13754

@article{Thomas2022-08pract-57910, title={A practical guide for generating unsupervised, spectrogram-based latent space representations of animal vocalizations}, year={2022}, doi={10.1111/1365-2656.13754}, number={8}, volume={91}, issn={0021-8790}, journal={Journal of Animal Ecology}, pages={1567--1581}, author={Thomas, Mara and Jensen, Frants H. and Averly, Baptiste and Demartsev, Vlad and Manser, Marta B. and Sainburg, Tim and Roch, Marie A. and Strandburg-Peshkin, Ariana} }

Strandburg-Peshkin, Ariana 2022-08 2022-07-01T09:18:15Z Sainburg, Tim Demartsev, Vlad Demartsev, Vlad Jensen, Frants H. Attribution-NonCommercial 4.0 International Averly, Baptiste Roch, Marie A. Jensen, Frants H. 2022-07-01T09:18:15Z Sainburg, Tim Manser, Marta B. Thomas, Mara Thomas, Mara Roch, Marie A. Strandburg-Peshkin, Ariana A practical guide for generating unsupervised, spectrogram-based latent space representations of animal vocalizations Averly, Baptiste eng 1. Background: The manual detection, analysis and classification of animal vocalizations in acoustic recordings is laborious and requires expert knowledge. Hence, there is a need for objective, generalizable methods that detect underlying patterns in these data, categorize sounds into distinct groups and quantify similarities between them. Among all computational methods that have been proposed to accomplish this, neighbourhood-based dimensionality reduction of spectrograms to produce a latent space representation of calls stands out for its conceptual simplicity and effectiveness.<br /><br />2. Goal of the study/what was done: Using a dataset of manually annotated meerkat Suricata suricatta vocalizations, we demonstrate how this method can be used to obtain meaningful latent space representations that reflect the established taxonomy of call types. We analyse strengths and weaknesses of the proposed approach, give recommendations for its usage and show application examples, such as the classification of ambiguous calls and the detection of mislabelled calls.<br /><br />3. What this means: All analyses are accompanied by example code to help researchers realize the potential of this method for the study of animal vocalizations. Manser, Marta B.

Downloads since Jul 1, 2022 (Information about access statistics)

Thomas_2-w1lzfbwes3iv8.pdf 43

This item appears in the following Collection(s)

Attribution-NonCommercial 4.0 International Except where otherwise noted, this item's license is described as Attribution-NonCommercial 4.0 International

Search KOPS


Browse

My Account