Professor of Software Engineering
I develop methods for mining animals sounds. Together with my biologist collaborators I have worked on sounds from diverse animals including whales, parrots, primates and bats.
Current Research Activities
Quantitative analysis of timing in animal vocal sequences
Project in collaboration with Christian Montes-Medina from UNAM, Mexico and Marc Timme from TU Dresden
Timing features such as the silence gaps between vocal units — inter-call intervals (ICIs) — often correlate with biological information such as context or genetic information. Such correlates between the ICIs and biological information have been reported for a diversity of animals. Yet, few quantitative approaches for investigating timing exist to date. Here, we propose a novel approach for quantitatively comparing timing in animal vocalisations in terms of the typical ICIs. As features, we use the distribution of silence gaps parametrised with a kernel density estimate (KDE) and compare the distributions with the symmetric Kullback-Leibler divergence (sKL-divergence). We use this technique to compare timing in vocalisations of two frog species, a group of zebra finches and calls from parrots of the same species.
Vocal sequences of Long-Finned Pilot Whales
In collaboration with S. Hallerberg, K. Hammerschmidt, M. Timme, H. Vester form HAW Hamburg, DPZ Göttingen, TU Dresden and Ocean Sounds.
Long-finned pilot whales are species of dolphins about which little is known. Like most delphinids, they are highly social and vocally flexible. Long-finned pilot whales produce stereotypic tonal sounds – known as calls – believed to play important social roles. Because of this, investigating how pilot whales communicate exchanging calls can shade light into the social organization of these animals. In this project, we investigate temporal and combinatorial patterns in call sequences of a group of pilot whales using non-parametric statistical methods.
Pylotwhale: A Python package for automatically annotating bioacoustic recordings
In collaboration with Marc Timme from TU Dresden
PylotWhale is a Python package for automatically annotating bioacoustic recordings by combining available tools for machine learning (Scikit-learn) and audio signal processing (Librosa). With Pylotwhale you can handle annotated audio files, extract audio features and transform classifier predictions into annotation files.
In collaboration with Alexander Kirschel from the University of Cyprus
Antthrush are a highly vocal species of Ground Antbird that communicate using individually distinctive sounds referred to as songs. Discerning an antthrush’s song can be valuable to understanding the species’ social interactions. Song detection is often done manually; yet, this approach limits the amount of data that can be processed and is prone to human biases and errors. It is also incredibly time consuming. Automating this step would be of enormous value for the study of antthrushes, and the method could be adapted for use with other species with similar song structure. The purpose of this project is to use supervised machine learning techniques to, first, detect antthrush songs and then identify the individuals singing. The model will be trained on a dataset of recordings with antthrush songs annotated manually. The data was collected in the south of Mexico by Dr. Kirschel from the University of Cyprus. Discerning individual species’ vocalisations can help us understand how individuals of a species interact vocally, monitor the species in the wild and support conservation efforts.
Florencia Noriega, Christian Montes-Medina, and Marc Timme. Quantitative analysis of timing in animal vocal sequences, 2019.
Andrea Ravignani, Simone Dalla Bella, Simone Falk, Christopher T Kello, Florencia Noriega, and Sonja A Kotz. Rhythm in speech and animal vocalizations: a cross-species perspective. Annals of the New York Academy of Sciences, 2019.
Invited Talks and Conferences
Bioacoustics – A Complex Multi-dimensional Bioacoustics View
Liminar Inverstigations of Memory and Brain Organization
Quantifying Timing in Animal Vocalizations
International Bioacoustics Congress