No abstract

Some aspects of the neural mechanisms underlying mouse ultrasonic vocalizations (USVs) are a useful model for the neurobiology of human speech and speech-related disorders. Much of the research on vocalizations and USVs is limited to offline methods and supervised classification of USVs, hindering the discovery of new types of vocalizations and the study of real-time free behavior. To address these issues, we developed AMVOC (Analysis of Mouse VOcal Communication) as a free, open-source software to analyze and detect USVs in both online and offline modes. When compared to hand-annotated ground-truth USV data, AMVOC's detection functionality (both offline and online) has high accuracy, and outperforms leading methods in noisy conditions, thus allowing for broader experimental use. AMVOC also includes the implementation of an unsupervised deep learning approach that facilitates discovery and analysis of USV data by clustering USVs using latent features extracted by a convolutional autoencoder and isimplemented in a graphical user interface (GUI), also enabling user's evaluation. These results can be used to explore the vocal repertoire space of the analyzed vocalizations. In this way, AMVOC will facilitate vocal analyses in a broader range of experimental conditions and allow users to develop previously inaccessible experimental designs for the study of mouse vocal behavior.