#eNeuro: Riekers et al. developed Moove (Marking Online using only the Onsets of Vocal Elements), an open-source, machine learning based tool, for real-time segmentation and classification of birdsong. https://doi.org/10.1523/ENEURO.0023-26.2026
Franziska Heubach
@franziheubach.bsky.social
Teaching birds since 2024 🎶​ 📍 Veit Lab, University of Tübingen
So excited to finally share this project🤗 Many thanks to my co-author Lena Veit! Check out our pre-print now to learn more about this new approach on song learning with positive reinforcement in songbirds! www.biorxiv.org/content/10.6...
biorxiv.org
Yet another step toward understanding song better🥳 Go check it out! www.biorxiv.org/content/10.6...
Syllable repetition reveals links between distant phrases in birdsong
Repeated execution of individual behavioural units is a common feature of many learned motor behaviours such as dance, music, and birdsong. Little is known about the neuronal control of such learned motor sequences, and specifically, how the number of variable repetitions is determined. The songs of Bengalese finches (Lonchura striata domestica) consist of individual syllables which can repeat a variable number of times (repeat number) to form a repeat phrase. Like vocal sequences in other animals, Bengalese finch song syntax is typically modelled as a Markov chain, where the choice to repeat the same syllable type or switch to a different one is made stochastically after each syllable, before the next syllable is produced. Here, we report that repeat numbers of adjacent and distant repeat phrases in the song can be correlated across specific pairs of phrases. These hidden links between distinct phrases challenge existing models of song syntax where the number of repetitions is independently determined for each syllable type. Instead, they suggest an organisation where a joint factor can control multiple nonadjacent phrases in a song. Repeat phrases in Bengalese finches may therefore be particularly suited to study the neuronal mechanisms underlying long-range dependencies in complex vocal sequences. ### Competing Interest Statement The authors have declared no competing interest. Deutsche Forschungsgemeinschaft, https://ror.org/018mejw64, 536953998, 532521431
biorxiv.org
Our python package for real-time targeting of birdsong syllables "Moove" is finally out! Thanks to my co-authors Nils Riekers, @goebl.bsky.social and Lena Veit. Our pre-print is online on biorxiv and our code is available on GitHub. Check it out! doi.org/10.64898/202... github.com/veitlab/moove
Real-Time Segmentation and Classification of Birdsong Syllables for Learning Experiments
Songbirds are essential animal models for studying neuronal and behavioral mechanisms of learned vocalizations. Bengalese finch (Lonchura striata domestica) songs contain a limited number of acoustica...
biorxiv.org