Katherine Zacarian

ORCID: 0009-0005-3780-7304
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Species Distribution and Climate Change
  • Animal Vocal Communication and Behavior
  • Marine animal studies overview
  • Bat Biology and Ecology Studies
  • Wildlife Ecology and Conservation
  • Music and Audio Processing
  • Environmental DNA in Biodiversity Studies

Earth Island Institute
2024

University of St Andrews
2023

New methods promise transformative insights and conservation benefits.

10.1126/science.adg7314 article EN Science 2023-07-13

The use of machine learning (ML) based techniques has become increasingly popular in the field bioacoustics over last years. Fundamental requirement for successful application ML are curated, agreed upon, high-quality datasets and benchmark tasks to be learned on a given dataset. However, so far lacks such public benchmarks which cover multiple species measure performance controlled standardized way that allows benchmarking newly proposed existing ones. Here, we propose BEANS (the BEnchmark...

10.1109/icassp49357.2023.10096686 article EN ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2023-05-05

Animal-borne sensors ('bio-loggers') can record a suite of kinematic and environmental data, which elucidate animal ecophysiology improve conservation efforts. Machine learning techniques are used for interpreting the large amounts data recorded by bio-loggers, but there exists no common framework comparing different machine in this domain. To address this, we present Bio-logger Ethogram Benchmark (BEBE), collection datasets with behavioral annotations, as well modeling task evaluation...

10.48550/arxiv.2305.10740 preprint EN cc-by arXiv (Cornell University) 2023-01-01

The use of machine learning (ML) based techniques has become increasingly popular in the field bioacoustics over last years. Fundamental requirements for successful application ML are curated, agreed upon, high-quality datasets and benchmark tasks to be learned on a given dataset. However, so far lacks such public benchmarks which cover multiple species measure performance controlled standardized way that allows benchmarking newly proposed existing ones. Here, we propose BEANS (the BEnchmark...

10.48550/arxiv.2210.12300 preprint EN cc-by arXiv (Cornell University) 2022-01-01
Coming Soon ...