Labeled frog-call dataset of Yasuní National Park for training Machine Learning algorithms
Published: 15 January 2019| Version 1 | DOI: 10.17632/5j852hzfjs.1
Contributors:
Museo de Zoología QCAZ, Andrés Estrella Terneux, Damián Nicolalde, Daniel Nicolalde, Samael PadillaDescription
Labeled dataset of frog calls recorded at Yasuní National Park using directional and omni-directional microphones. We trained a Bayesian classifier with Gaussian Mixture Models and applied it to study the audio in Unidentified Long Recordings. The results of the analysis are in the related article. We showed that the application of a frog-call recognition algorithm allows the estimation of presence-absence of the trained subset of species in real-world recordings of an actual amphibian monitoring made in the wild.
Files
Steps to reproduce
Apply the algorithm described in the related article.
Institutions
Pontificia Universidad Catolica del Ecuador
Categories
Amphibians, Machine Learning, Bioacoustics, Soundscapes, Classifier Evaluation