Assessing the Potential of Birdnet to Infer European Bird Communities from Large-Scale Ecoacoustic Data DOI
David Funosas, Luc Barbaro, Laura Schillé

и другие.

Опубликована: Янв. 1, 2023

1. Passive acoustic monitoring has become increasingly popular as a practical and cost-effective way of obtaining highly reliable data in ecological research projects. Increased ease collecting these means that, currently, the main bottleneck ecoacoustic projects is often time required for manual analysis passively collected recordings. In this study we evaluate potential current limitations BirdNET-Analyzer v2.4, most advanced generic deep learning algorithm bird recognition to date, tool assess community composition through automated large-scale data. 2. To end, 3 datasets comprising total 629 environmental soundscapes 194 different sites spread across 19° latitude span Europe. We analyze recordings both with BirdNET by listening local expert birders, then compare results obtained two methods performance at level each single vocalization entire recording sequences (1, 5 or 10 min). 3. Our analyses reveal that identifications can be if sufficiently high minimum confidence threshold used. However, recall markedly low when adjusted ensure levels precision. Thus, found F1-scores remain moderate (<0.5) all thresholds studied. therefore estimate extended duration are currently necessary provide minimally comprehensive picture target community. also suggest not significantly influenced type recorder used habitat recorded but modulated volume species-specific available online. 4. conclude judicious use AI-based IDs provided represent novel powerful method assist assessment Finally, best recommendations optimal from communities.

Язык: Английский

Assessing the potential of BirdNET to infer European bird communities from large-scale ecoacoustic data DOI Creative Commons
David Funosas, Luc Barbaro, Laura Schillé

и другие.

Ecological Indicators, Год журнала: 2024, Номер 164, С. 112146 - 112146

Опубликована: Май 20, 2024

Passive acoustic monitoring has become increasingly popular as a practical and cost-effective way of obtaining highly reliable data in ecological research projects. Increased ease collecting these means that, currently, the main bottleneck ecoacoustic projects is often time required for manual analysis passively collected recordings. In this study we evaluate potential current limitations BirdNET-Analyzer v2.4, most advanced generic deep learning algorithm bird recognition to date, tool assess community composition through automated large-scale data. To end, 3 datasets comprising total 629 environmental soundscapes 194 different sites spread across 19° latitude span Europe. We analyze using both BirdNET listening by local expert birders, then compare results obtained two methods performance at level each single vocalization entire recording sequences (1, 5 or 10 min). Since provides confidence score identification, minimum thresholds can be used filter out identifications with low scores, thus retaining only ones. The volume did not allow us estimate species-specific taxa, so instead evaluated global selected optimized when consistently applied all species. Our analyses reveal that if sufficiently high threshold used. However, inevitable trade-off between precision recall does obtain satisfactory metrics same time. found F1-scores remain moderate (<0.5) studied, extended duration seem currently necessary provide minimally comprehensive picture target community. estimate, however, usage species- context-specific would substantially improve benchmarks study. conclude judicious use AI-based provided represent powerful method assist assessment data, especially duration.

Язык: Английский

Процитировано

14
Avian vocalizations in Huangmaohai sea-crossing channel: Automatic birdsong recognition and ecological impact analysis based on deep learning DOI
Tiandou Hu, Minmin Yuan,

Jinhui Li

и другие.

Biological Conservation, Год журнала: 2025, Номер 305, С. 111101 - 111101

Опубликована: Март 25, 2025

Язык: Английский

Процитировано

0
A dataset of acoustic measurements from soundscapes collected worldwide during the COVID-19 pandemic DOI Creative Commons
Samuel Challéat, Nicolas Farrugia, Jérémy S. P. Froidevaux

и другие.

Scientific Data, Год журнала: 2024, Номер 11(1)

Опубликована: Авг. 27, 2024

Political responses to the COVID-19 pandemic led changes in city soundscapes around globe. From March October 2020, a consortium of 261 contributors from 35 countries brought together by Silent Cities project built unique soundscape recordings collection report on local acoustic urban areas. We present this here, along with metadata including observational descriptions areas contributors, open-source environmental data, confinement levels and calculation descriptors. performed technical validation dataset using statistical models run subset manually annotated soundscapes. Results confirmed large-scale usability ecoacoustic indices automatic sound event recognition collection. expect be useful for research multidisciplinary field sciences.

Язык: Английский

Процитировано

1
Assessing the potential of BirdNET to infer European bird communities from large-scale ecoacoustic data DOI Creative Commons
David Funosas, Luc Barbaro, Laura Schillé

и другие.

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2023, Номер unknown

Опубликована: Дек. 12, 2023

Abstract Passive acoustic monitoring has become increasingly popular as a practical and cost-effective way of obtaining highly reliable data in ecological research projects. Increased ease collecting these means that, currently, the main bottleneck ecoacoustic projects is often time required for manual analysis passively collected recordings. In this study we evaluate potential current limitations BirdNET-Analyzer v2.4, most advanced generic deep learning algorithm bird recognition to date, tool assess community composition through automated large-scale data. To end, 3 datasets comprising total 629 environmental soundscapes 194 different sites spread across 19° latitude span Europe. We analyze recordings both with BirdNET by listening local expert birders, then compare results obtained two methods performance at level each single vocalization entire recording sequences (1, 5 or 10 min). Our analyses reveal that identifications can be if sufficiently high minimum confidence threshold used. However, recall markedly low when adjusted ensure levels precision. Thus, found F1-scores remain moderate (<0.5) all thresholds studied. therefore estimate extended duration are currently necessary provide minimally comprehensive picture target community. also suggest not significantly influenced type recorder used habitat recorded but modulated volume species-specific available online. conclude judicious use AI-based IDs provided represent novel powerful method assist assessment Finally, best recommendations optimal from communities.

Язык: Английский

Процитировано

1
Assessing the Potential of Birdnet to Infer European Bird Communities from Large-Scale Ecoacoustic Data DOI
David Funosas, Luc Barbaro, Laura Schillé

и другие.

Опубликована: Янв. 1, 2023

1. Passive acoustic monitoring has become increasingly popular as a practical and cost-effective way of obtaining highly reliable data in ecological research projects. Increased ease collecting these means that, currently, the main bottleneck ecoacoustic projects is often time required for manual analysis passively collected recordings. In this study we evaluate potential current limitations BirdNET-Analyzer v2.4, most advanced generic deep learning algorithm bird recognition to date, tool assess community composition through automated large-scale data. 2. To end, 3 datasets comprising total 629 environmental soundscapes 194 different sites spread across 19° latitude span Europe. We analyze recordings both with BirdNET by listening local expert birders, then compare results obtained two methods performance at level each single vocalization entire recording sequences (1, 5 or 10 min). 3. Our analyses reveal that identifications can be if sufficiently high minimum confidence threshold used. However, recall markedly low when adjusted ensure levels precision. Thus, found F1-scores remain moderate (<0.5) all thresholds studied. therefore estimate extended duration are currently necessary provide minimally comprehensive picture target community. also suggest not significantly influenced type recorder used habitat recorded but modulated volume species-specific available online. 4. conclude judicious use AI-based IDs provided represent novel powerful method assist assessment Finally, best recommendations optimal from communities.

Язык: Английский

Процитировано

1