Abstract:Vocal communication is a major component of animal social behavior. Vocalizations can be learned or innate, and can convey a variety of signals, including territorial limits, the presence of predators, or courtship intent. Mouse ultrasonic vocalizations (USVs) are a promising model in which to study mammalian vocal production circuits. While mouse USVs are innate, mice still show complex vocal behavior, including production of structured song composed of multiple syllable types and the ability to modify their vocal rate and syllable repertoire based on social conditions. Though in courtship interactions male mice produce the majority of the emitted USVs, female mice are capable of emitting USVs. In order to study the underlying mechanisms of vocal production in freely behaving pairs of mice, it is necessary to identify the individual responsible for each syllable in group settings. Previous methods to identify the source of an individual USVs have used high-density microphone arrays and triangulation methods, which involve the use of multiple costly microphones and require implementation of complex computational methods. Here we identified, developed, and used an inexpensive, mountable, ultrasound-sensitive miniature-microphone system to record and identify USVs from individual mice during dual socializing behavior. Our system includes custom circuit boards that can be fitted to individual mice and connected to a variety of existing USV recording systems. We found that these miniature microphones reliably detected mouse USVs, and that a high percentage (90%) of vocalizations could be attributed to a specific animal in a vocalizing pair based on the relative amplitude differences alone. This simple readout method avoids the implementation of complicated triangulation methods. By pairing this method with simultaneous video recording and automated animal body part and identity tracking, we were able to study and describe the broader courtship behavioral landscape, in which USV production is one component. These results offer a promising, low-cost, and simple method that researchers can implement to study the social vocal communication between at least pairs of vocalizing mice.

Utilizing DeepSqueak for automatic detection and classification of mammalian vocalizations: a case study on primate vocalizations

Introducing a Central African Primate Vocalisation Dataset for Automated Species Classification

Development of deep neural networks for marine mammal call detection using an open-source, user friendly tool

Utilizing synthetic training data for the supervised classification of rat ultrasonic vocalizations

ANALYSIS OF ULTRASONIC VOCALIZATIONS FROM MICE USING COMPUTER VISION AND MACHINE LEARNING

Automated detection of Bornean white-bearded gibbon (Hylobates albibarbis) vocalizations using an open-source framework for deep learning

Applying machine learning to primate bioacoustics: Review and perspectives

Automated detection of Bornean white-bearded gibbon (Hylobates albibarbis) vocalisations using an open-source framework for deep learning

Automatic Sound Event Detection and Classification of Great Ape Calls Using Neural Networks

Enhancing the analysis of murine neonatal ultrasonic vocalizations: Development, evaluation, and application of different mathematical models

Automatic detection and classification of marmoset vocalizations using deep and recurrent neural networks.

Automated Call Detection for Acoustic Surveys with Structured Calls of Varying Length

Same data, different results? Evaluating machine learning approaches for individual identification in animal vocalisations

Detection and classification of vocal productions in large scale audio recordings

Deep Machine Learning Techniques for the Detection and Classification of Sperm Whale Bioacoustics

HybridMouse: A Hybrid Convolutional-Recurrent Neural Network-Based Model for Identification of Mouse Ultrasonic Vocalizations

A workflow for the automated detection and classification of female gibbon calls from long-term acoustic recordings

MUPET—Mouse Ultrasonic Profile ExTraction: A Signal Processing Tool for Rapid and Unsupervised Analysis of Ultrasonic Vocalizations

Mountable miniature microphones to identify and assign mouse ultrasonic vocalizations

Learning to rumble: Automated elephant call classification, detection and endpointing using deep architectures