Automated acoustic analysis of 50-kHz ultrasonic vocalizations using template matching and contour analysis

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

Bioacoustic analyses of animal vocalizations are predominantly accomplished through manual scanning, a highly subjective and time-consuming process. Thus, validated automated analyses are needed that are usable for a variety of animal species and easy to handle by non-programing specialists. This study tested and validated whether DeepSqueak, a user-friendly software, developed for rodent ultrasonic vocalizations, can be generalized to automate the detection/segmentation, clustering and classification of high-frequency/ultrasonic vocalizations of a primate species. Our validation procedure showed that the trained detectors for vocalizations of the gray mouse lemur (Microcebus murinus) can deal with different call types, individual variation and different recording quality. Implementing additional filters drastically reduced noise signals (4225 events) and call fragments (637 events), resulting in 91% correct detections (N_total = 3040). Additionally, the detectors could be used to detect the vocalizations of an evolutionary closely related species, the Goodman’s mouse lemur (M. lehilahytsara). An integrated supervised classifier classified 93% of the 2683 calls correctly to the respective call type, and the unsupervised clustering model grouped the calls into clusters matching the published human-made categories. This study shows that DeepSqueak can be successfully utilized to detect, cluster and classify high-frequency/ultrasonic vocalizations of other taxa than rodents, and suggests a validation procedure usable to evaluate further bioacoustics software.

Acoustilytix™: A Web-Based Automated Ultrasonic Vocalization Scoring Platform

Ultrasonic vocalizations (USVs) are known to reflect emotional processing, brain neurochemistry, and brain function. Collecting and processing USV data is manual, time-intensive, and costly, creating a significant bottleneck by limiting researchers’ ability to employ fully effective and nuanced experimental designs and serving as a barrier to entry for other researchers. In this report, we provide a snapshot of the current development and testing of Acoustilytix™, a web-based automated USV scoring tool. Acoustilytix implements machine learning methodology in the USV detection and classification process and is recording-environment-agnostic. We summarize the user features identified as desirable by USV researchers and how these were implemented. These include the ability to easily upload USV files, output a list of detected USVs with associated parameters in csv format, and the ability to manually verify or modify an automatically detected call. With no user intervention or tuning, Acoustilytix achieves 93% sensitivity (a measure of how accurately Acoustilytix detects true calls) and 73% precision (a measure of how accurately Acoustilytix avoids false positives) in call detection across four unique recording environments and was superior to the popular DeepSqueak algorithm (sensitivity = 88%; precision = 41%). Future work will include integration and implementation of machine-learning-based call type classification prediction that will recommend a call type to the user for each detected call. Call classification accuracy is currently in the 71–79% accuracy range, which will continue to improve as more USV files are scored by expert scorers, providing more training data for the classification model. We also describe a recently developed feature of Acoustilytix that offers a fast and effective way to train hand-scorers using automated learning principles without the need for an expert hand-scorer to be present and is built upon a foundation of learning science. The key is that trainees are given practice classifying hundreds of calls with immediate corrective feedback based on an expert’s USV classification. We showed that this approach is highly effective with inter-rater reliability (i.e., kappa statistics) between trainees and the expert ranging from 0.30–0.75 (average = 0.55) after only 1000–2000 calls of training. We conclude with a brief discussion of future improvements to the Acoustilytix platform.

Rodent ultrasonic vocalizations as biomarkers of future alcohol use: A predictive analytic approach

Excessive alcohol consumption has a vast, negative impact on society. Rodent models have been successful in furthering our understanding of the biological underpinnings that drive alcohol consumption. Rodents emit ultrasonic vocalizations (USVs) that are each composed of several acoustic characteristics (e.g., frequency, duration, bandwidth, power). USVs reflect neurotransmitter activity in the ascending limb of the mesolimbic dopaminergic and cholinergic neurotransmitter systems and serve as noninvasive, real-time biomarkers of dopaminergic and cholinergic neurotransmission in the limbic system. In the present study, we recorded spontaneously emitted USVs from alcohol-naïve Long-Evans (LE) rats and then measured their alcohol intake. We compared the USV acoustic characteristics and alcohol consumption data from these LE rats with previously published data from selectively bred high-alcohol (P and HAD-1) and low-alcohol (NP and LAD-1) drinking lines from studies with the same experimental method. Predictive analytic techniques were applied simultaneously to this combined data set and revealed that (a) USVs emitted by alcohol-naïve rats accurately discriminated among high-alcohol consuming, LE, and low-alcohol consuming rat lines, and (b) future alcohol consumption in these same rat lines was reliably predicted from the USV data collected in an alcohol-naïve state. To our knowledge, this is the first study to show that alcohol consumption is predicted directly from USV profiles of alcohol-naïve rats. Because USV acoustic characteristics are sensitive to underlying neural activity, these findings suggest that baseline differences in mesolimbic cholinergic and dopaminergic tone could determine the propensity for future alcohol consumption in rodents.

Sex-specific ultrasonic vocalization patterns and alcohol consumption in high alcohol-drinking (HAD-1) rats

Ultrasonic vocalizations (USVs) have been established as an animal model of emotional status and are often utilized in drug abuse studies as motivational and emotional indices. Further USV functionality has been demonstrated in our recent work showing accurate identification of selectively-bred high versus low alcohol-consuming male rats ascertained exclusively from 22 to 28kHz and 50-55kHz FM USV acoustic parameters. With the hypothesis that alcohol-sensitive sex differences could be revealed through USV acoustic parameters, the present study examined USVs and alcohol consumption in male and female selectively bred high-alcohol drinking (HAD-1) rats. For the current study, we examined USV data collected during a 12-week experiment in male and female HAD-1 rats. Experimental phases included Baseline (2weeks), 4-h EtOH Access (4weeks), 24-h EtOH Access (4weeks) and Abstinence (2weeks). Findings showed that both male and female HAD-1 rats spontaneously emitted a large number of 22-28kHz and 50-55kHz FM USVs and that females drank significantly more alcohol compared to males over the entire course of the experiment. Analyses of USV acoustic characteristics (i.e. mean frequency, duration, bandwidth and power) revealed distinct sex-specific phenotypes in both 50-55kHz FM and 22-28kHz USV transmission that were modulated by ethanol exposure. Moreover, by using a linear combination of these acoustic characteristics, we were able to develop binomial logistic regression models able to discriminate between male and female HAD-1 rats with high accuracy. Together these results highlight unique emotional phenotypes in male and female HAD-1 rats that are differentially modulated by alcohol experience.