Overview of BirdCLEF 2022: Endangered bird species recognition in soundscape recordings

As the “extinction capital of the world”, Hawai‘i has lost 68% of its native bird species, the consequences of which can harm entire ecosystems. With physical monitoring difficult, scientists have turned to sound recordings, as this approach could provide a passive, low labor, and cost-effective strategy for monitoring endangered bird populations. Current methods for processing large bioacoustic datasets involve manual review of each recording. This requires specialized training and prohibitively large amounts of time. Recent advances in machine learning have made it possible to automatically identify bird songs for common species with ample training data. However, it remains challenging to develop such tools for rare and endangered species. The main goal of the 2022 edition of BirdCLEF was to advance automated detection of rare and endangered bird species that lack large amounts of training data. The competition challenged participants to develop reliable analysis frameworks to detect and identify the vocalizations of rare bird species in continuous Hawaiian soundscapes utilizing limited training data.

[1]  Marcos V. Conde,et al.  Few-shot Long-Tailed Bird Audio Recognition , 2022, CLEF.

[2]  Anthony Miyaguchi,et al.  Motif Mining and Unsupervised Representation Learning for BirdCLEF 2022 , 2022, CLEF.

[3]  Silvia Zuffi,et al.  Perspectives in machine learning for wildlife conservation , 2021, Nature Communications.

[4]  Christof Henkel,et al.  Recognizing bird species in diverse soundscapes under weak supervision , 2021, CLEF.

[5]  Connor M. Wood,et al.  Survey coverage, recording duration and community composition affect observed species richness in passive acoustic surveys , 2020, Methods in Ecology and Evolution.

[6]  Xiaobai Liu,et al.  Deep neural networks for automated detection of marine mammal species , 2020, Scientific Reports.

[7]  Diego Llusia,et al.  A roadmap for survey designs in terrestrial acoustic monitoring , 2019, Remote Sensing in Ecology and Conservation.

[8]  P. Hart,et al.  Loss of cultural song diversity and the convergence of songs in a declining Hawaiian forest bird community , 2019, Royal Society Open Science.

[9]  Eamonn J. Keogh,et al.  Fast Similarity Matrix Profile for Music Analysis and Exploration , 2019, IEEE Transactions on Multimedia.

[10]  Diego Llusia,et al.  Terrestrial Passive Acoustic Monitoring: Review and Perspectives , 2018, BioScience.

[11]  S. Kendall,et al.  Research and management priorities for Hawaiian forest birds , 2018, The Condor.

[12]  Stefano Ermon,et al.  Tile2Vec: Unsupervised representation learning for spatially distributed data , 2018, AAAI.

[13]  E. Martynov,et al.  Dealing with Class Imbalance in Bird Sound Classification , 2022, CLEF.

[14]  D. Kowerko,et al.  TUC Media Computing at BirdCLEF 2022: Strategies in identifying bird sounds in a complex acoustic environments , 2022, CLEF.

[15]  Rhythm Garg,et al.  Bird Species Classification: One Step at a Time , 2022, CLEF.

[16]  Stefan Kahl,et al.  BirdNET: A deep learning solution for avian diversity monitoring , 2021, Ecol. Informatics.

[17]  Bernd Freisleben,et al.  Bird Species Recognition via Neural Architecture Search , 2020, CLEF.

[18]  Mario Lasseck,et al.  Bird Species Identification in Soundscapes , 2019, CLEF.

[19]  Jan Schlüter,et al.  Bird Identification from Timestamped, Geotagged Audio Recordings , 2018, CLEF.

[20]  T. Pratt Conservation biology of Hawaiian forest birds : implications for island avifauna , 2009 .