APLOSE: A web-based annotation platform for underwater passive acoustic monitoring
Gabriel Dubus,
Maëlle Torterotot,
Julie Béesau,
Mathieu Dupont,
Anatole Gros-Martial,
Mathilde Michel,
Elodie Morin,
Paul Nguyen Hong Duc,
Pierre-Yves Raumer,
Olivier Adam,
Flore Samaran,
Dorian Cazau
Affiliations
Gabriel Dubus
Sorbonne University, CNRS, Institut d'Alembert UMR 7190, LAM, Paris, France; ENSTA Bretagne Lab-STICC UMR CNRS 6285, Brest, France; Corresponding author at: Institut Jean le Rond d'Alembert Lutheries Acoustique Musique, France.
Maëlle Torterotot
ENSTA Bretagne Lab-STICC UMR CNRS 6285, Brest, France
Julie Béesau
ENSTA Bretagne Lab-STICC UMR CNRS 6285, Brest, France
Mathieu Dupont
ENSTA Bretagne Lab-STICC UMR CNRS 6285, Brest, France
Anatole Gros-Martial
ENSTA Bretagne Lab-STICC UMR CNRS 6285, Brest, France; Centre of Biological Studies of Chizé, CNRS, Chizé, France
Mathilde Michel
ENSTA Bretagne Lab-STICC UMR CNRS 6285, Brest, France
Elodie Morin
ENSTA Bretagne Lab-STICC UMR CNRS 6285, Brest, France
Paul Nguyen Hong Duc
Centre for Marine Science and Technology, Curtin University, Perth, Australia
Pierre-Yves Raumer
ENSTA Bretagne Lab-STICC UMR CNRS 6285, Brest, France; Geo-Ocean, Univ Brest, CNRS, Ifremer, UMR 6538 and ENSTA Bretagne Lab-STICC UMR CNRS 6285, Brest, France
Olivier Adam
Sorbonne University, CNRS, Institut d'Alembert UMR 7190, LAM, Paris, France
Flore Samaran
ENSTA Bretagne Lab-STICC UMR CNRS 6285, Brest, France
Dorian Cazau
ENSTA Bretagne Lab-STICC UMR CNRS 6285, Brest, France
Emerging detection and classification algorithms based on deep learning models require manageable large-scale manual annotations of ground truth data. To date, the challenge of creating large and accurate annotated datasets of underwater sounds has been a major obstacle to the development of robust recognition algorithms. APLOSE (Annotation PLatform for Ocean Sound Explorers) is an open-source, web-based tool which facilitates collaborative annotation campaigns in underwater acoustics. The platform was used to carry out research projects on inter-annotator variability, to build training and testing data sets for detection algorithms and to perform bioacoustics analysis on noisy datasets. In the future, it will enable the creation of high-quality reference datasets to test and train the new detection and classification algorithms.