Search Results (196)

Automated pig-welfare monitoring needs scalable, non-invasive signals that work across ages and individuals. A key methodological contribution of this study is the use of subject-wise validation, which ensures generalization to unseen animals and prevents inflated accuracy caused by growth-related and individual ‘voice’ differences. Vocalizations can help, but growth and individual “voice” differences can confound distress patterns and overstate accuracy without subject-wise validation. In our study, we explicitly accounted for individual variability by including animal identity as a random effect in mixed models and by using grouped cross-validation, where models were tested only on pigs not seen during training. This approach ensures that the reported accuracy reflects generalization across different individuals rather than memorization of specific vocal signatures. We analyzed 2221 vocal samples from 40 pigs (20 males, 20 females) recorded across four growth phases (farrowing, nursery, growing, finishing) under six conditions (pain, hunger, thirst, cold stress, heat stress, normal). Acoustic features extracted in Praat included energy, duration, intensity, pitch, and formants (F1–F4). Using blockwise variance decomposition, we quantified contributions of distress exposure, growth phase, and sex, and estimated the additional variance explained by animal identity. Distress exposure dominated intensity and spectral traits, particularly Formant 2, whereas the growth phase produced systematic shifts in duration and pitch. Animal identity added a modest but consistent increment in explained variance (~+0.02–0.03 R² beyond sex, phase, and distress). For prediction, we used 5-fold cross-validation grouped by animal. A Random Forest achieved a modest balanced accuracy of 0.609 and macro-F1 of 0.597; pain was most separable (recall 0.825), while other states showed moderate recall, indicating overlap. These results support hierarchical acoustic encoding of distress and establish a benchmark for precision welfare monitoring. Furthermore, they highlight that resolving complex physiological overlaps, such as heat stress and resource competition, requires a shift from unimodal acoustic models to multimodal Precision Livestock Farming (PLF) systems that integrate bioacoustics with continuous environmental and behavioral data streams. Full article

Urban wetlands are acoustic hotspots where vegetation structure, hydrological dynamics, and anthropogenic noise interact, yet multi-season assessments of how vegetation influences avian soundscapes are limited. This study explored bird soundscape dynamics across forest, open forest grassland, and meadow habitats in Nanjing Xinjizhou National Wetland Park, eastern China, using passive acoustic monitoring during spring and autumn 2023. Twelve sampling points (four per vegetation type) were established, and six acoustic indices were calculated, including the Acoustic Complexity Index (ACI), Acoustic Diversity Index (ADI), Acoustic Evenness Index (AEI), Bioacoustic Index (BIO), Normalized Difference Soundscape Index (NDSI), and Acoustic Entropy Index (H). were calculated from 48-h recordings each season. Random forest models and redundancy analysis assessed the relationships between acoustic indices, fine-scale vegetation parameters (e.g., crown width, tree height, species richness), and anthropogenic factors (e.g., distance to roads/trails, surface hardness). Vegetation structure, particularly crown width, was the primary driver of avian acoustic diversity, with broad-crowned forests consistently exhibiting the highest acoustic complexity. In spring, anthropogenic factors such as trail and road proximity dominated soundscape variation, suppressing biological sounds. In autumn, with reduced human presence, vegetation structure emerged as the dominant factor, while bioacoustic activity remained elevated despite reduced peaks in acoustic complexity. Proximity to roads increased low-frequency (1–2 kHz) noise and suppressed mid-frequency (4–8 kHz) bird vocalizations, but trees with crown widths ≥4 m maintained higher acoustic diversity even near disturbance sources. This study demonstrates that vegetation structure mediates both resource availability and sound propagation, buffering the effects of anthropogenic disturbance in frequency-specific ways. Multi-season sampling is crucial for understanding the dynamic interplay between vegetation phenology and human activity that shapes urban wetland soundscapes. Full article

Swim bladders in some teleost fish can act as gas-filled cavities that oscillate under acoustic pressure and transfer the sound energy to the inner ears. Quantifying the resonance frequency and damping of these oscillations is useful for linking swim bladder mechanics to hearing-related and behavioural questions, but many established direct-measure approaches have relied on open-water deployments and careful avoidance of boundary reflections, making experiments logistically demanding and difficult to reproduce (e.g., requiring deep-water sites, careful control of surface/boundary reflections, and complex deployment geometries). This study presents a compact laboratory methodology for estimating swim bladder resonance properties using a closed, fully water-filled, stainless-steel impedance tube. Broadband pseudorandom excitation is applied via an end-plate shaker, and the acoustic response of the system is recorded using wall-mounted hydrophones. Resonance peaks are identified using power spectral estimates of recorded signals, allowing resonance frequency and quality factor to be extracted from the peak location and −3 dB bandwidth. The approach is first established using inflated latex balloons as surrogate encapsulated gas cavities, providing a controlled benchmark for repeatability and interpretation. It is then applied to recently euthanised brown trout (Salmo trutta), where clear resonance features attributable to the swim bladder are observed and show systematic variation with body size. A coupled finite element model reproduces the principal resonance behaviour under the experimental loading and supports interpretation of the measured peaks as swim bladder resonance. The results provide a validated foundation for subsequent non-invasive measurements on live, free-swimming fish, as well as for future applications where swim bladder condition may be relevant to management or conservation. Full article

Urban parks derived from historical forest fragments represent important refugia for biodiversity in rapidly expanding cities. The wood cricket, Nemobius sylvestris, was surprisingly found in a park in the northern part of Bucharest, Romania, an area under exponential residential development. The species was confirmed by calling song analysis and molecularly confirmed through DNA-barcoding. The acoustic analysis revealed substantial geographic variation in the signals of N.sylvestris across its European range, with the Romanian population exhibiting the most distinctive acoustic characteristics. A median joining network was constructed using available COI sequences from public databases, showing moderate genetic variability within European samples. This flightless, woodland-specialist cricket is highly sensitive to habitat fragmentation and its persistence in this urban park demonstrates the conservation value of retaining semi-natural forest structure within city green spaces. Our findings highlight the importance of urban parks as biodiversity refugia, particularly for habitat specialists with limited dispersal abilities. This discovery underscores the need for the integrative conservation management of urban forest remnants, emphasizing the retention of natural structural elements such as leaf litter and heterogeneous canopy cover to support diverse invertebrate communities. Full article

The interaction between humans and birds plays an important role in shaping the sustainability of tropical rainforest ecosystems, particularly through bird vocalizations that function as bioacoustic indicators of ecological conditions while simultaneously embedding socio-cultural meanings within local communities. This study aims to (1) classify types and categories of bird sounds as perceived by rural communities, and (2) assess the role of bird vocalizations as cultural symbols supporting community-based conservation practices. The study was conducted across six islands and eight villages in North Maluku, Eastern Indonesia, using a qualitative approach based on semi-structured interviews and community workshops. A total of 435 respondents, all of whom were farmers residing along forest margins, participated in the study. The results documented 51 bird species belonging to 26 families, whose vocalizations were interpreted and classified by local communities into three acoustic categories: 21 species with loud calls (41.18%), 12 species with melodious calls (23.53%), and 18 species with sad calls (35.29%). Melodious vocalizations were commonly associated with values of beauty, calmness, and social harmony, whereas loud calls were predominantly interpreted as warnings, signals of alertness, or indicators of environmental change. These findings demonstrate that bird sounds serve not merely as ecological cues, but as culturally embedded symbols that guide daily activities, moral values, and conservation ethics within rural communities. By documenting the cultural significance of bird vocalizations across a clearly defined geographic context, this study provides an empirical basis for culturally informed conservation strategies aimed at protecting bird species subject to high levels of cultural use and ecological pressure. Full article

The Greater Caribbean manatee faces significant conservation challenges due to a lack of demographic data in low-visibility habitats. To address this, we present a refined automated manatee counting method pipeline integrating deep learning-based call detection with unsupervised individual counting. We resolved significant computational bottlenecks by implementing an offline feature extraction strategy, bypassing a 13-h processing lag for 43,031 audio samples. To mitigate overfitting in imbalanced bioacoustic datasets, non-parametric bootstrap resampling was employed to generate 100,000 balanced spectrograms. Benchmarking revealed that transfer learning via a VGG-16 backbone achieved a mean 10-fold cross-validation accuracy of 98.92% (±0.08%) and an F1-score of 98.08% for genuine vocalizations. Following detection, individual counting utilized k-means clustering on prioritized music information retrieval descriptors—spectral bandwidth, centroid, and roll-off—to resolve distinct acoustic signatures. This framework identified three individuals with a silhouette coefficient of 79.20%, demonstrating superior cohesion over previous benchmarks. These results confirm the automatic manatee count method as a robust, scalable framework for generating the scientific evidence required for regional conservation policies. Full article

Deep neural networks are widely used for image classification in different fields, although selecting an appropriate architecture often remains a trial-and-error process. The purpose of this work is to investigate a convolutional neural network architecture used to detect whale pulses in spectrograms in order to better understand the causes of its underperformance. By examining the behaviour of its internal layers, we show that the early convolutional blocks capture the most informative acoustic features, while deeper layers provide limited additional benefit and, under the considered training conditions, may even degrade classification accuracy. Based on these observations, we derive a simplified architecture consisting of only the first two convolutional layers followed by a lightweight classifier. This network achieves near-optimal performance, improving accuracy from 87% to 98%, and exhibits substantially lower variability between repetitions compared to the original model. Full article

The 1st International Online Conference on Taxonomy (IOCTX2025) serves as a critical interdisciplinary nexus for addressing the contemporary “taxonomic impediment” through the lens of integrative systematics and computational innovation. By synthesizing research spanning from Paleozoic fossil records to extant microbial biodiversity, the conference illuminates the evolving methodology of species delimitation, moving beyond traditional morphometrics to incorporate multi-locus molecular phylogenetics, bioacoustics, and high-resolution 3D imaging. Key thematic clusters across the program examine the floristic complexity of Karst landscapes, the resolution of cryptic animal species complexes through genomic and proteomic data, and the role of machine learning in automating the identification of both fossil and living taxa. Furthermore, the proceedings underscore a paradigm shift toward “integrative taxonomy,” where the fusion of morphological rigor with eDNA metabarcoding and automated genomic scanning provides a more robust framework for understanding global biodiversity hotspots. Ultimately, IOCTX2025 reaffirms taxonomy as a high-technology discipline essential for conservation biology and evolutionary theory, providing a standardized scientific language to describe the complexities of the tree of life across deep time and modern ecosystems. Full article

Anurans are among the most threatened vertebrates worldwide, yet their acoustic ecology in fragmented habitats remains understudied. This research investigated acoustic overlaps and resource partitioning among amphibian species inhabiting Maceira Pond in Caparaó National Park, Brazil using bioacoustic methods. Six hours of recordings were analysed to determine key acoustic parameters and identify the resident species. A principal component analysis was used to assess acoustic parameters, whilst a cluster analysis examined acoustic similarities. Twelve species from four families were detected, of which eight were identified and five remained unidentified. Four species showed over 90% acoustic overlap, while two had less than 50%, with one at about 17%. Central frequency, peak frequency, duration, bandwidth, and pace significantly contributed to call differentiation. The R-value confirmed clustering patterns, indicating likely low acoustic interference due to few sympatric species. This study provides the first acoustic niche assessment for this community and highlights the need for further research on spatial and temporal partitioning in these threatened amphibian assemblages. Full article

The rapid global expansion of offshore wind energy (OWE) has established it as a critical component of the renewable energy transition; however, this development concurrently introduces significant underwater noise pollution into marine ecosystems. This paper provides a comprehensive review of the acoustic footprint of OWE across its entire lifecycle, rigorously distinguishing between the high-intensity, acute impulsive noise generated during pile-driving construction and the chronic, low-frequency continuous noise associated with decades-long turbine operation. We critically evaluate the engineering capabilities and limitations of current underwater acoustic monitoring architectures, including buoy-based real-time monitoring nodes, cabled high-bandwidth systems (e.g., cabled hydrophone arrays with DAQ/DSP and fiber-optic distributed acoustic sensing, DAS), and autonomous seabed archival recorders (PAM deployment). Furthermore, documented biological impacts are synthesized across diverse taxa, ranging from auditory masking and threshold shifts in marine mammals to the often-overlooked sensitivity of invertebrates and fish to particle motion—a key metric frequently missing from standard pressure-based assessments. Our analysis identifies a fundamental gap in current governance paradigms, which disproportionately prioritize the mitigation of short-term acute impacts while neglecting the cumulative ecological risks of long-term operational noise. This review synthesizes recent evidence on chronic operational noise and outlines a conceptual pathway from event-based compliance monitoring toward long-term, adaptive soundscape management. We propose the implementation of integrated, adaptive acoustic monitoring networks capable of quantifying cumulative noise exposure and informing real-time mitigation strategies. Such a paradigm shift is essential for optimizing mitigation technologies and ensuring the sustainable coexistence of marine renewable energy development and marine biodiversity. Full article

Interspecies communication is increasingly recognized as an affective–cognitive process co-created between humans and animals rather than a one-directional transmission of signals. This review integrates findings from ethology, neuroscience, welfare science, behavioral studies, and posthumanist ethics to examine how emotional expression, communicative intentionality, and relational engagement shape understanding across species. Research on primates, dogs, elephants, and marine mammals demonstrates that empathy, consolation, cooperative signaling, and multimodal perception rely on evolutionarily conserved mechanisms, including mirror systems, affective contagion, and oxytocin-mediated bonding. These biological insights intersect with ethical considerations concerning animal agency, methodological responsibility, and the interpretation of non-human communication. Emerging technological tools—bioacoustics, machine vision, and AI-assisted modeling—offer new opportunities to analyze complex vocal and behavioral patterns, yet they require careful contextualization to avoid anthropocentric misclassification. Synthesizing these perspectives, the review proposes a relational framework in which meaning arises through shared emotional engagement, embodied interaction, and ethically grounded interpretation. This approach highlights the importance of welfare-oriented, minimally invasive methodologies and supports a broader shift toward recognizing animals as communicative partners whose emotional lives contribute to scientific knowledge. This review primarily synthesizes empirical and theoretical research on primates and dogs, complemented by selected examples from elephants and marine mammals, which provide the most developed evidence base for the affective–cognitive and relational mechanisms discussed. Full article

Passive acoustic monitoring (PAM) using hydrophones enables collecting acoustic data to be collected in large and diverse quantities, necessitating the need for a reliable automated classification system. This paper presents a data-processing pipeline and a set of neural networks designed for a humpback-whale-detection system. A collection of audio segments is compiled using publicly available audio repositories and extensively curated via manual methods, undertaking thorough examination, editing and clipping to produce a dataset minimizing bias or categorization errors. An array of standard data-augmentation techniques are applied to the collected audio, diversifying and expanding the original dataset. Multiple neural networks are designed and trained using TensorFlow 2.20.0 and Keras 3.13.1 frameworks, resulting in a custom curated architecture layout based on research and iterative improvements. The pre-trained model MobileNetV2 is also included for further analysis. Model performance demonstrates a strong dependence on both feature representation and network architecture. Mel spectrogram inputs consistently outperformed MFCC (Mel-Frequency Cepstral Coefficients) features across all model types. The highest performance was achieved by the pretrained MobileNetV2 using mel spectrograms without augmentation, reaching a test accuracy of 99.01% with balanced precision and recall of 99% and a Matthews correlation coefficient of 0.98. The custom CNN with mel spectrograms also achieved strong performance, with 98.92% accuracy and a false negative rate of only 0.75%. In contrast, models trained with MFCC representations exhibited consistently lower robustness and higher false negative rates. These results highlight the comparative strengths of the evaluated feature representations and network architectures for humpback whale detection. Full article

This work presents the first study addressing two-dimensional numerical simulations of acoustic wave scattering involving a simplified human body model placed inside an enclosed cabin. The simulations utilise the µ-diff backscattering algorithm in MATLAB, which is suitable for modeling frequency-domain interactions with multiple scatterers under penetrable boundary conditions. The body is represented as a cluster of penetrable, tangent circular cylinders with acoustic properties mimicking muscle, fat, bone, and clothing layers. Hidden PVC cylinders are embedded to simulate concealed objects. Several configurations were examined, varying the number of PVC inclusions (two to four), the frequency range, and the presence of an absorbing cabin wall. Sound pressure level (SPL) distributions around the body and at a 1 m distance were analysed. Polar plots reveal distinct differences between the baseline body model and those incorporating PVC inclusions. The most pronounced effects occur near 160 Hz, where an absorbing wall is present within the acoustic enclosure. The presence of an absorbing wall modifies wave behaviour, producing enhanced directional attenuation. The results demonstrate how object composition, spatial arrangement, and enclosure geometry influence acoustic backscattered fields. These findings highlight the potential of wave-based numerical modelling for detecting concealed items on the human body in confined acoustic environments, supporting the development of non-invasive security screening technologies. Full article

Analyzing large datasets of animal vocalizations requires efficient bioacoustic methods for categorizing the vocalization types. This study evaluates the effectiveness of different vocalization clustering methods, comparing fully automated and semi-automated methods against the gold standard of manual expert annotations. Effective methods achieve good clustering performance whilst minimizing human effort. We release a new dataset of 1454 zebra finch vocalizations manually clustered by experts, on which we evaluate (i) fully automated clustering using off-the-shelf methods based on sound embeddings and (ii) a semi-automated workflow relying on refining the embedding-derived clusters. Clustering performance is assessed using the Adjusted Rand Index (ARI) and Normalized Mutual Information (NMI). Results indicate that while fully automated methods provide a useful baseline, they generally fall short of human-level consistency. In contrast, the semi-automated workflow achieved agreement scores comparable to inter-expert reliability, approaching the levels of expert manual clustering. This demonstrates that refining embedding-derived clusters reduces annotation time while maintaining gold standard accuracy. We conclude that semi-automated workflows offer an optimal strategy for bioacoustics, enabling the scalable analysis of large datasets without compromising the precision required for robust behavioral insights. Full article

Rising numbers of wolf (Canis lupus) populations make traditional, resource-intensive methods of wolf monitoring increasingly challenging and often insufficient. This study explores how wolf howls can be used as a new monitoring tool for wolves by applying Artificial Intelligence (AI) methods to detect and classify wolf howls from acoustic recordings, thereby improving the effectiveness of wolf population monitoring. Three AI approaches are evaluated: BirdNET, Yellowstone’s Cry-Wolf project system, and BioLingual. Data were collected using Song Meter SM4 (SM4) audio recorders in a known wolf territory in Klelund Dyrehave, Denmark, and manually validated to establish a ground truth of 260 wolf howls. Results demonstrate that while AI solutions currently do not achieve the complete precision or overall accuracy of expert manual analysis, they offer tremendous efficiency gains, significantly reducing processing time. BirdNET achieved the highest recall at 78.5% (204/260 howls detected), though with a low precision of 0.007 (resulting in 28,773 false positives). BioLingual detected 61.5% of howls (160/260) with 0.005 precision (30,163 false positives), and Cry-Wolf detected 59.6% of howls (155/260) with 0.005 precision (30,099 false positives). Crucially, a combined approach utilizing all three models achieved a 96.2% recall (250/260 howls detected). This suggests that while AI solutions primarily function as powerful human-aided data reduction tools rather than fully autonomous detectors, they represent a valuable, scalable, and non-invasive complement to traditional methods in wolf research and conservation, making large-scale monitoring more feasible. Full article