Fine-Tuning for Bird Sound Classification: An Empirical Study
Abstract
In biodiversity research scientists now routinely acquire audio recordings of vocalizing bird species and are then faced with the task of identifying the species audible in these recordings. Here, we analyze the accuracy (precision, recall and $F_1$ F 1 score) of several deep networks, in conjunction with pre-training and data augmentation techniques, for classifying audio recordings of twelve bird species under multiple data scarcity settings.
Cite
Text
Stein and Andres. "Fine-Tuning for Bird Sound Classification: An Empirical Study." European Conference on Computer Vision Workshops, 2024. doi:10.1007/978-3-031-92387-6_15Markdown
[Stein and Andres. "Fine-Tuning for Bird Sound Classification: An Empirical Study." European Conference on Computer Vision Workshops, 2024.](https://mlanthology.org/eccvw/2024/stein2024eccvw-finetuning/) doi:10.1007/978-3-031-92387-6_15BibTeX
@inproceedings{stein2024eccvw-finetuning,
title = {{Fine-Tuning for Bird Sound Classification: An Empirical Study}},
author = {Stein, David and Andres, Bjoern},
booktitle = {European Conference on Computer Vision Workshops},
year = {2024},
pages = {198-207},
doi = {10.1007/978-3-031-92387-6_15},
url = {https://mlanthology.org/eccvw/2024/stein2024eccvw-finetuning/}
}