AVN: A Deep Learning Approach for the Analysis of Birdsong.

Deep learning tools for behavior analysis have enabled important new insights and discoveries in neuroscience. Yet, they often compromise interpretability and generalizability for performance, making it difficult to quantitively compare phenotypes across datasets and research groups. We developed a novel deep learning-based behavior analysis pipeline, Avian Vocalization Network (AVN), for the learned vocalizations of the most extensively studied vocal learning model species - the zebra finch. AVN annotates songs with high accuracy across multiple animal colonies without the need for any additional training data and generates a comprehensive set of interpretable features to describe the syntax, timing, and acoustic properties of song. We use this feature set to compare song phenotypes across multiple research groups and experiments, and to predict a bird's stage in song development. Additionally, we have developed a novel method to measure song imitation that requires no additional training data for new comparisons or recording environments, and outperforms existing similarity scoring methods in its sensitivity and agreement with expert human judgements of song similarity. These tools are available through the open-source AVN python package and graphical application, which makes them accessible to researchers without any prior coding experience. Altogether, this behavior analysis toolkit stands to facilitate and accelerate the study of vocal behavior by enabling a standardized mapping of phenotypes and learning outcomes, thus helping scientists better link behavior to the underlying neural processes.

Expression of FoxP2 in the basal ganglia regulates vocal motor sequences in the adult songbird.

Disruption of the transcription factor FoxP2, which is enriched in the basal ganglia, impairs vocal development in humans and songbirds. The basal ganglia are important for the selection and sequencing of motor actions, but the circuit mechanisms governing accurate sequencing of learned vocalizations are unknown. Here, we show that expression of FoxP2 in the basal ganglia is vital for the fluent initiation and termination of birdsong, as well as the maintenance of song syllable sequencing in adulthood. Knockdown of FoxP2 imbalances dopamine receptor expression across striatal direct-like and indirect-like pathways, suggesting a role of dopaminergic signaling in regulating vocal motor sequencing. Confirming this prediction, we show that phasic dopamine activation, and not inhibition, during singing drives repetition of song syllables, thus also impairing fluent initiation and termination of birdsong. These findings demonstrate discrete circuit origins for the dysfluent repetition of vocal elements in songbirds, with implications for speech disorders.