COSS
Mis à jour le 22/03/2024Catalogue of sound signatures
Objectives
The aim of this project is to build an open library of sound signatures from the coastal seascape of European waters. This library will serve as a learning and testing base for open source artificial intelligence algorithms for the automatic classification of signals of anthropogenic origin.
The objectives of this European project will be to set up tools to support the implementation of the MSFD (EU Maritime Strategy Framework Directive) for Descriptor 11 dedicated to underwater noise. It will provide means of determining the acoustic impact of human activities and identify mitigation measures to manage these activities so that they do not harm the marine environment.
Shom contribution
The Shom is heavily involved in three of the project's tasks, which bring together the following three themes:
- Definition of a standard format for acoustic signatures;
- Exploitation of raw recordings (signal annotation);
- Exploitation of the tools developed in the project to support public policy and the implementation of European legislation.
In addition, this project is part of a more general drive to increase Shom's (DASM's) competence in a number of areas:
- Support for our in-house classification work, with the possibility of exchanges with other experts in the field:
- Database of annotated signal ;
- Labeling of acoustic data;
- Analysis methods.
- Increased expertise in AI processing of acoustic data, thanks to the context and collaboration with AI experts.
Duration
18 months – from 07/2023 to 12/2024
Funding program
European Climate, Infrastructure and Environment Executive Agency (CINEA) - Project Call: CALL - CINEA/2022/OP/0019
Partnership
The project is being carried out by a consortium comprising the following partners in addition to Shom:
- Centro Tecnológico Naval y del Mar (CTNaval, Spain) - coordinator
- ICES (International Council for the Exploration of the Sea, Denmark)
- ETT (Systems and big data, Italy)
- SAES Electronica Submarine (SMA and electronics, Spain)
- OnAIR (IA development and data modeling, Italy)
As well as the following subcontractors:
- Witteveen+Bos (Engineering and consulting, Netherlands) ;
- SMHI Swedish Meteorological and Hydrological Institute (Sweden) ;
- Aarhus University (Denmark).
Project contacts
Project manager at Shom : Laura Ceyrac