Overview of BirdCLEF 2019: Large-Scale Bird Recognition in Soundscapes

Stefan Kahl (1) , Fabian-Robert Stöter (2) , Hervé Goëau (3) , Hervé Glotin (4) , Robert Planque (5) , Willem-Pier Vellinga (6) , Alexis Joly (2)

1 Chemnitz University of Technology / Technische Universität Chemnitz
2 ZENITH - Scientific Data Management
3 UMR AMAP - Botanique et Modélisation de l'Architecture des Plantes et des Végétations
4 UTLN - Université de Toulon
5 Department of Mathematics [Amsterdam]
6 Xeno-canto foundation

The BirdCLEF challenge-as part of the 2019 LifeCLEF Lab [7]-offers a large-scale proving ground for system-oriented evaluation of bird species identification based on audio recordings. The challenge uses data collected through Xeno-canto, the worldwide community of bird sound recordists. This ensures that BirdCLEF is close to the conditions of real-world application, in particular with regard to the number of species in the training set (659). In 2019, the challenge was focused on the difficult task of recognizing all birds vocalizing in omni-directional soundscape recordings. Therefore, the dataset of the previous year was extended with more than 350 hours of manually annotated soundscapes that were recorded using 30 field recorders in Ithaca (NY, USA). This paper describes the methodology of the conducted evaluation as well as the synthesis of the main results and lessons learned.

Mots clés: Ecological monitoring; Collection; Audio; Species; Retrieval; Call; Song; Bird; LifeCLEF; Identification; Fine-grained classification; Evaluation; Bioacoustics; Benchmark

Domaines: Systématique, phylogénie et taxonomie; Ecosystèmes; Botanique; Biodiversité et Ecologie; Informatique [cs]

Stefan Kahl, Fabian-Robert Stöter, Hervé Goëau, Hervé Glotin, Robert Planque, et al.. Overview of BirdCLEF 2019: Large-Scale Bird Recognition in Soundscapes. CLEF 2019 - Conference and Labs of the Evaluation Forum, Cappellato, L.; Ferro, N.; Losada, D. E.; Müller, H., Sep 2019, Lugano, Switzerland. ⟨hal-02345644⟩