Master thesis »Deep Learning-Based Audio Analysis for Avi...

Intro

The Fraunhofer Institute for Digital Media Technology IDMT offers a master's thesis opportunity focused on deep learning-based audio analysis for estimating avian populations.

Tasks

Explore and develop audio analysis methods for counting bird populations in outdoor environments.
Develop deep learning-based audio analysis methods to count distinct bird calls in complex soundscapes.
Investigate suitable bird song datasets with polyphony annotations or create a dataset from existing recordings.
Identify and re-implement two deep learning-based approaches for bird counting.
Train and evaluate models for bird counting using the compiled dataset.
Test the robustness of the counting method in real-world scenarios with passive acoustic monitoring sensors.
Simulate various microphone characteristics and evaluate their effect on counting accuracy.
Document results in a written thesis.

Requirements

Excellent skills in audio signal processing and deep learning.
Practical experience using Python and deep learning libraries such as TensorFlow or PyTorch.
General interest in bioacoustic research topics.

Benefits

Exciting market-related topics with complex issues to solve.
Excellent opportunities for professional and technical training.
Space to implement own ideas in quarterly open-topic contests.
Excellent technical infrastructure.
Collaboration with renowned partners and customers.
Very good work-life balance with flexible working hours and mobile working options.
Open-minded and interested team with a tolerant atmosphere and regular team events.
Good transport connections and proximity to the state capital Erfurt.
Attractive special offers as part of Fraunhofer corporate benefits.

Master thesis »Deep Learning-Based Audio Analysis for Avian Population Estimation«

Intro

Tasks

Requirements

Benefits

Similar impact jobs