Research Engineer/Post-doctorate position in deep learning for marine megafauna monitoring from acoustic data (F/M) – Warning: please send your application directly and only by e-mail to

  • Reference : FEM-SAS-2022-324
  • Position type: Fixed-term contract
  • Duration: 12 month
  • Localisation: Brest (29) - France Energies Marines et ENSTA Bretagne, Vannes (56) Université de Bretagne Sud
  • Download the offer

Job Description

The post doctorate will work in the Environmental Integration of ORE Program and contribute to the OWFSOMM project, which aims to standardise tools for monitoring marine megafauna at the scale of offshore windfarms. Offshore wind has emerged as one of the most dynamic technologies in the energy mix and is rapidly expanding. Offshore windfarms may have impacts on marine megafauna, impacts that have to be estimated and reduced in the context of environmental policies. For any offshore wind farm, monitoring of marine megafauna is thus crucial at each step from consenting to construction and operation. Environmental platforms at sea are being increasingly developed and deployed to monitor various environmental parameters, together with monitoring the marine megafauna. Each sensor has its own limitations and data are usually post-processed with the help of data catalogues used to label  and identify species, behaviour, etc. With the significant breakthrough of Artificial Intelligence, especially with deep neural networks (e.g. recurrent and/or convolutional), automatization of these monitoring processes have become a realistic objective as well as crucial to deal with continuous recordings and large datasets.

This R&D proposal is part of the OWFSOMM project (Offshore Wind Farm Surveys of Marine Megafauna: standardization of tools and methods for monitoring at OWF scales) among several academic research laboratories and industrial partners. One of the work-packages of this project is to develop deep learning frameworks for automatic detection of megafauna using multimodal data with the final aim of estimating high-level ecological indicators. Among several data sources, underwater passive acoustic signals appear to be the most significant ones that should be exploited within the automatic detection framework.

Preliminary works have been conducted to develop deep networks for marine mammal detection based on spectrograms generated from acoustic data. Current results show that this approach is promising and needs deeper investigation to provide more fruitful results. Therefore, the objective of this proposal is to design effective deep learning models that could improve the detection and classification of marine megafauna. In this context, advanced deep models based on supervised, semi-supervised  as well as self-supervised learning will be considered to adopt the best solution.


In order to address the afore mentioned objectives, a tentative work program is given below.

  • Bibliographical study of deep learning-based methods for marine mammal detection and classification using spectrograms from acoustic data.
  • Evaluation and benchmarking of state-of-the-art methods.
  • Improvement of existing solutions (including the current in-house methods developed by the team) and development of new models based on supervised/semi-supervised learning approaches.
  • Dissemination: recommendation report to the ORE sector, publication, source codes.

The successful candidate will join the team Environmental Integration of ORE of France Energies Marines in Brest, France. He/She will be jointly supervised by Dr. Karine Heerah (Permanent researcher at France Energies Marines, coordinator of the OWFSOMM project), Dr. Dorian Cazau (Associate professors at ENSTA Bretagne) and Dr.  Minh-Tan Pham (Associate professor at Université Bretagne-Sud/IRISA).

Required Skills

Initial training
PhD/M.Sc/M.Eng in Signal Processing or Computer Science or related

Work experience
• Experience in signal and image processing, applied machine learning
• Experience with acoustic data, spectrogrammetry
• Work experience and knowledge on marine megafauna

Specific knowledge
• Excellent programming skills in Python (familiar with one of deep learning packages, such as PyTorch or Tensorflow, is a must)
• Interest for applied environmental issues
• Knowledge of environmental and conservation issues in the context of the European Marine Directive Strategy Framework and ORE

Professional assets
• Great scientific rigor
• Spirit of initiative and multidisciplinary openness
• Taste for applied research (industrial research)
• Ease of expression, argumentation, and communication in a partnership context
• Taste for teamwork but ability to work autonomously as well

This position is open to people with disabilities.

Warning: due to issues with the website, please send your application directly and only by e-mail to


Job : Research Engineer/Post-doctorate position in deep learning for marine megafauna monitoring from acoustic data (F/M) – Warning: please send your application directly and only by e-mail to

Reference : FEM-SAS-2022-324

    Closed search overlay screen