If anybody needs another book recommendation, The Blue Sweater by Jacqueline Novogratz is extremely well written and gives insights in how to battle poverty with a small markets approach.

Do you have recordings from multiple microphones? Or do you just want to run a simple call detection algorithm or trace the fundamental frequency? Our new R package might be able to do just that. Read the preprint on BioRxiv: https://doi.org/10.1101/2023.02.07.527470. Any questions are very welcome either here, by email or as an issue on GitHub. Thanks to Stephen Tyndel, Barbara Klump, Gustavo Alarcon-Nieto and Lucy Aplin for all their hard work!

Thanks to Juan Carlos Senar, @LucyAplin and Mary Brooke McElreath for all their hard work. Thanks to MPI-AB, Centre for the Adv Study of Collective Behaviour and IMPRS QBEE for supporting this work.

Do #parrots know who is talking? It took some time, but you can read the answer in our preprint on parrot vocal signatures and voice-prints: https://doi.org/10.1101/2023.01.20.524864. We find evidence that monk parakeets have a #voice, which is recognisable across call types! Very happy that my first paper on the original subject of my PhD is finally being submitted.

Three years ago I started writing code to calculate flamingo flock sizes throughout time together with Johanna Staerk. I had no idea why, but it was a fun challenge. More than three years later it all makes sense. Andrew Mooney and others have figured out how many flamingoes you need to optimise breeding and help conserve these amazing species! Read more here: https://doi.org/10.1002/zoo.21753

Do you want to classify multiple call types? No problem, just run the multi-class version of the network and it should do just that. For any questions don’t hesitate to contact Christian Bergler or me. Thanks to all co-authors, IMPRS QBEE, MPI AB and all funding agents! To Barbara Klump and Nora Carlson for organising the initial workshop! And of course first author Christian Bergler, who developed the network and made this all happen!

Do you have many hours of audio data? Do simple call detection algorithms not work well enough? Don’t despair, after more than three years of development Animal-SPOT is here! Published in Scientific Reports: https://doi.org/10.1038/s41598-022-26429-y. All code and example data is publicly available. You can train your own network and detect vocalisations of anything from cod to whale and bat to parrot. #MachineLearning #Bioacoustics #DeepLearning