Animal vocalizations serve a wide range of functions including territorial defense, courtship, social cohesion, begging, and vocal learning. Whereas many insights have been gained from observational studies and experiments using auditory stimulation, there is currently no technology available for the selective control of vocal communication in small animal groups. We developed a system for real-time control of vocal interactions among separately housed animals. The system is implemented on a field-programmable gate array (FPGA) and it allows imposing arbitrary communication networks among up to four animals. To minimize undesired transitive sound leakage, we adopted echo attenuation and sound squelching algorithms. In groups of three zebra finches, we restrict vocal communication in circular and in hierarchical networks and thereby mimic complex eavesdropping and middleman situations.

Annotating and proofreading data sets of complex natural behaviors such as vocalizations are tedious tasks because instances of a given behavior need to be correctly segmented from background noise and must be classified with minimal false positive error rate. Low-dimensional embeddings have proven very useful for this task because they can provide a visual overview of a data set in which distinct behaviors appear in different clusters. However, low-dimensional embeddings introduce errors because they fail to preserve distances; and embeddings represent only objects of fixed dimensionality, which conflicts with vocalizations that have variable dimensions stemming from their variable durations. To mitigate these issues, we introduce a semi-supervised, analytical method for simultaneous segmentation and clustering of vocalizations. We define a given vocalization type by specifying pairs of high-density regions in the embedding plane of sound spectrograms, one region associated with vocalization onsets and the other with offsets. We demonstrate our two-neighborhood (2N) extraction method on the task of clustering adult zebra finch vocalizations embedded with UMAP. We show that 2N extraction allows the identification of short and long vocal renditions from continuous data streams without initially committing to a particular segmentation of the data. Also, 2N extraction achieves much lower false positive error rate than comparable approaches based on a single defining region. Along with our method, we present a graphical user interface (GUI) for visualizing and annotating data.