Automatic detection and classification of bearded seal vocalizations in the northeastern Chukchi Sea using convolutional neural networks

EMD-Based Energy Spectrum Entropy Distribution Signal Detection Methods for Marine Mammal Vocalizations

To develop a passive acoustic monitoring system for diversity detection and thereby adapt to the challenges of a complex marine environment, this study harnesses the advantages of empirical mode decomposition in analyzing nonstationary signals and introduces energy characteristics analysis and entropy of information theory to detect marine mammal vocalizations. The proposed detection algorithm has five main steps: sampling, energy characteristics analysis, marginal frequency distribution, feature extraction, and detection, which involve four signal feature extraction and analysis algorithms: energy ratio distribution (ERD), energy spectrum distribution (ESD), energy spectrum entropy distribution (ESED), and concentrated energy spectrum entropy distribution (CESED). In an experiment on 500 sampled signals (blue whale vocalizations), in the competent intrinsic mode function (IMF2) signal feature extraction function distribution of ERD, ESD, ESED, and CESED, the areas under the curves (AUCs) of the receiver operating characteristic (ROC) curves were 0.4621, 0.6162, 0.3894, and 0.8979, respectively; the Accuracy scores were 49.90%, 60.40%, 47.50%, and 80.84%, respectively; the Precision scores were 31.19%, 44.89%, 29.44%, and 68.20%, respectively; the Recall scores were 42.83%, 57.71%, 36.00%, and 84.57%, respectively; and the F1 scores were 37.41%, 50.50%, 32.39%, and 75.51%, respectively, based on the threshold of the optimal estimated results. It is clear that the CESED detector outperforms the other three detectors in signal detection and achieves efficient sound detection of marine mammals.

There You Are! Automated Detection of Indris' Songs on Features Extracted from Passive Acoustic Recordings

The growing concern for the ongoing biodiversity loss drives researchers towards practical and large-scale automated systems to monitor wild animal populations. Primates, with most species threatened by extinction, face substantial risks. We focused on the vocal activity of the indri (Indri indri) recorded in Maromizaha Forest (Madagascar) from 2019 to 2021 via passive acoustics, a method increasingly used for monitoring activities in different environments. We first used indris’ songs, loud distinctive vocal sequences, to detect the species’ presence. We processed the raw data (66,443 10-min recordings) and extracted acoustic features based on the third-octave band system. We then analysed the features extracted from three datasets, divided according to sampling year, site, and recorder type, with a convolutional neural network that was able to generalise to recording sites and previously unsampled periods via data augmentation and transfer learning. For the three datasets, our network detected the song presence with high accuracy (>90%) and recall (>80%) values. Once provided the model with the time and day of recording, the high-performance values ensured that the classification process could accurately depict both daily and annual habits of indris‘ singing pattern, critical information to optimise field data collection. Overall, using this easy-to-implement species-specific detection workflow as a preprocessing method allows researchers to reduce the time dedicated to manual classification.

Using transfer learning with a convolutional neural network to detect African manatee (Trichechus senegalensis) vocalizations

African manatees (Trichechus senegalensis) are vulnerable, understudied, and difficult to detect. Areas where African manatees are found were acoustically sampled and deep learning techniques were used to develop the first African manatee vocalization detector. A transfer learning approach was used to develop a convolutional neural network (CNN) using a pretrained CNN (GoogLeNet). The network was highly successful, even when applied to recordings collected from a different location. Vocal detections were more common at night and tended to occur within less than 2 min of one another.

Convolutional Neural Networks for the Identification of African Lions from Individual Vocalizations

The classification of vocal individuality for passive acoustic monitoring (PAM) and census of animals is becoming an increasingly popular area of research. Nearly all studies in this field of inquiry have relied on classic audio representations and classifiers, such as Support Vector Machines (SVMs) trained on spectrograms or Mel-Frequency Cepstral Coefficients (MFCCs). In contrast, most current bioacoustic species classification exploits the power of deep learners and more cutting-edge audio representations. A significant reason for avoiding deep learning in vocal identity classification is the tiny sample size in the collections of labeled individual vocalizations. As is well known, deep learners require large datasets to avoid overfitting. One way to handle small datasets with deep learning methods is to use transfer learning. In this work, we evaluate the performance of three pretrained CNNs (VGG16, ResNet50, and AlexNet) on a small, publicly available lion roar dataset containing approximately 150 samples taken from five male lions. Each of these networks is retrained on eight representations of the samples: MFCCs, spectrogram, and Mel spectrogram, along with several new ones, such as VGGish and stockwell, and those based on the recently proposed LM spectrogram. The performance of these networks, both individually and in ensembles, is analyzed and corroborated using the Equal Error Rate and shown to surpass previous classification attempts on this dataset; the best single network achieved over 95% accuracy and the best ensembles over 98% accuracy. The contributions this study makes to the field of individual vocal classification include demonstrating that it is valuable and possible, with caution, to use transfer learning with single pretrained CNNs on the small datasets available for this problem domain. We also make a contribution to bioacoustics generally by offering a comparison of the performance of many state-of-the-art audio representations, including for the first time the LM spectrogram and stockwell representations. All source code for this study is available on GitHub.

Introduction to the special issue on ocean acoustics in the changing arctic

This paper introduces the Special Issue of The Journal of the Acoustical Society of America on Ocean Acoustics in the Changing Arctic. The special issue includes papers on ocean (and in one case atmospheric) acoustics. Changes in both the ice cover and ocean stratification have significant implications for acoustic propagation and ambient sound. The Arctic is not done changing, and papers in this special issue, therefore, represent a snapshot of current acoustic conditions in the Arctic.