Neural networks that locate and identify birds through their songs
Departamento de Física, FCEyN, Universidad de Buenos Aires, Buenos Aires, Argentina
2 IFIBA, CONICET, Buenos Aires, Argentina
Accepted: 16 December 2021
Published online: 28 December 2021
In this work, we present a set of algorithms that allow the location and identification of birds through their songs. To achieve the first objective, neural networks capable of reconstructing the position of the subject are trained from a set of differences in the arrival times of a sound signal to the different microphones in an array. For the second objective, a dynamical system is used to generate surrogate songs, similar to those of a given set of subjects, to train a neural network so that it can classify subjects. Taken together, they constitute an interesting tool for the automatic monitoring of small bird populations.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2021