Master's thesis; Identifying Guillemot’s individuals through parent-chick vocal interaction
Master's thesis; Using AI to identify Guillemot’s individuals through parent-chick vocal interaction analysis
Background
The AukLab-Audio project represents a cutting-edge, long-term data collection effort at the Stora Karlsö Auk Lab, a unique cliff-edge observatory for common guillemots (Uria aalge) in the Baltic Sea. This initiative is generating an unprecedented multi-channel, multi-modal dataset, featuring continuous, synchronized audio recordings from an array of high-quality DPA microphones and camera-integrated microphones, alongside extensive video footage, weather data, and other sensor information. The richness of this dataset opens up unparalleled opportunities for research at the intersection of bioacoustics, machine learning, and marine ecology.
We are excited to offer four interconnected Master's thesis projects designed to explore different facets of this unique dataset. These projects aim to foster a collaborative research environment, where students will contribute distinct expertise towards a unified goal: to deepen our understanding of guillemot vocal behavior, develop novel computational analysis techniques for complex acoustic scenes, and derive new ecological insights. A potential research field trip to the Stora Karlsö research site may be organized to provide students with first-hand experience of the data collection environment and the study species.
Description
Guillemots look and sound all very similar to human senses, and our only way to distinguish between them is by attaching numbered rings to their legs. However, guillemots do recognize each other as they can find their chicks and partner within a colony through their calls. In this project we aim to take the first step ever towards uncovering their vocal uniqueness, what makes individual guillemots vocalization special?
In this project we will explore the potential of machine learning techniques to identify individual guillemots based on their vocalizations. Voice biometrics are typically used to analyze a human’s unique vocal characteristics and perform speaker or singer automatic recognition. Here we will translate these methodologies to Guillemots identification and evaluate their potential for individual recognition. We will first explore existing audio feature extraction techniques for human voice and later design one tailored for Guillemots vocal range and style. We will then explore different models for the identification of individuals.
We will focus on parent-chick interaction. Both parents, female and male, have equal role at the upbringing of their chicks on the cliff edge. When one incubates or stays with the chick, the other fishes and vice-versa. At some point, when the chick is of age, the parents encourage it to jump of the cliff towards the sea. This dramatic jump takes a lot of convincing, and an unique vocal exchange between parent and chick takes place before, during and after. Thanks to the video cameras, we can locate when this exchange has happened and use the audio recordings at that time as our input data for analysis.
Key Responsibilities
- Data handling and processing, to select and prepare the data for analysis
- Literature review on audio features for vocal biometrics and speaker/singer detection
- Implementation of relevant audio features and design of features tailored to guillemots
- Implementation of model for guillemots individual identification, both traditional ML and NNs can be considered
- Writing and defending master thesis
- Recurrent presentations of project progress
Qualifications
Required skills:
- Knowledge of audio signal processing
- Good programming skills in Python
- Experience in implementing machine learning models
- Interest in bioacoustics or animal behavior
Preferred skills:
- Experience with vocal biometrics
- Experience with speaker recognition or singer recognition
- Experience with audio features
Terms
- Location: Lund
- Time: January to June 2026
- Credits: 30 ECTS
- Compensation: In line with RISE guidelines for strategically important projects, a compensation of 30,000 SEK per student is offered upon completion and approval of the 30-credit thesis.
What we offer
- Supervision by an interdisciplinary team of researchers with expertise in machine learning, bioacoustics, and marine ecology.
- Access to the unique, large-scale AukLab-Audio dataset and computational resources.
- An opportunity to contribute to cutting-edge research with potential for high-impact publications.
- A collaborative and stimulating research environment.
- Potential for a research field trip to the Stora Karlsö field station.
Supervisors
- Delia Fano Yela, Dr (RISE)
- Jonas Hentati Sundberg, Dr (SLU)
- Sepideh Pashami, Dr (RISE)
Welcome with your application!
Last day of application: 23rd November 2025
Contact: delia.fano.yela@ri.se