OpenSoundscape: An open‐source bioacoustics analysis package for Python

Lapp, Sam; Rhinehart, Tessa; Freeland‐Haynes, Louis; Khilnani, Jatin; Syunkova, Alexandra; Kitzes, Justin

doi:10.1111/2041-210x.14196

Cited by 15 publications

(2 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the time required to classify camera‐trap images and audio data can be limiting, the implementation of efficient sampling schedules (Cifuentes et al., 2021) and the rapid evolution of CNNs and tools for implementing them are facilitating their wider use in conservation and ecological applications (Lapp et al., 2022; Vélez et al., 2023). For example, using a solid‐state drive and a central processing unit with 24 cores, we obtained CNN model predictions for classification of 800,000 audio clips in approximately 24 h. We encourage others to consider the combined use of these measurement technologies to assess how space is used by wildlife and disturbance factors in human‐dominated landscapes.…”

Section: Discussionmentioning

confidence: 99%

“…These recordings and their pattern of acoustic energy and frequency of sounds visible on the spectrograms were inspected by trained technicians and experts to classify the recordings as containing sounds of cattle, gunshots, or dogs or containing some other sound. These classifications were used to train species‐specific CNNs with OpenSoundscape 0.6.2 (Lapp et al., 2022), an open‐source Python utility library that provides machine learning tools for automated audio classification. We trained 2‐class, single‐species models to predict the presence or absence of cattle, gunshots, and dogs in the recordings collected in all sampling periods.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Implications of the scale of detection for inferring co‐occurrence patterns from paired camera traps and acoustic recorders

Vélez,

McShea,

Pukazhenthi

et al. 2024

Conservation Biology

View full text Add to dashboard Cite

Multifunctional landscapes that support economic activities and conservation of biological diversity (e.g., cattle ranches with native forest) are becoming increasingly important, as small remnants of native forest may comprise the only habitat left for some wildlife species. Understanding the co‐occurrence between wildlife and disturbance factors, such as poaching activity and domestic ungulates, is key to successful management of multifunctional landscapes. Tools to measure co‐occurrence between wildlife and disturbance factors include camera traps and autonomous acoustic recording units. We paired 52 camera trap stations with acoustic recorders to investigate the association between two measures of disturbance (poaching and cattle) and wild ungulates present in multifunctional landscapes of the Colombian Orinoquía. We used joint species distribution models to investigate species‐habitat associations and species‐disturbance correlations. One model was fitted using camera trap data to detect wild ungulates and disturbance factors, and a second model was fitted after replacing camera‐trap detections of disturbance factors with their corresponding acoustic detections. The direction, significance, and precision of the effect of covariates depended on the sampling method used for disturbance factors. Acoustic monitoring typically resulted in more precise estimates of the effects of covariates and of species‐disturbance correlations. Association patterns between wildlife and disturbance factors were only found when disturbance was detected by acoustic recorders. Camera traps allowed us to detect non‐vocalizing species whereas audio recording devices increased detection of disturbance factors leading to more precise estimates of co‐occurrence patterns. The collared peccary, lowland tapir and white‐tailed deer co‐occurred with disturbance factors and are conservation priorities due to greater risk of poaching or disease transmission from cattle.Article impact statement: Co‐occurrence of disturbance and wildlife depends on method of detection (camera traps or acoustic recorders).This article is protected by copyright. All rights reserved

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Implications of the scale of detection for inferring co‐occurrence patterns from paired camera traps and acoustic recorders

Vélez,

McShea,

Pukazhenthi

et al. 2024

Conservation Biology

View full text Add to dashboard Cite

show abstract

Performance of unmarked abundance models with data from machine‐learning classification of passive acoustic recordings

Fiss,

Lapp,

Cohen

et al. 2024

Ecosphere

View full text Add to dashboard Cite

The ability to conduct cost‐effective wildlife monitoring at scale is rapidly increasing due to the availability of inexpensive autonomous recording units (ARUs) and automated species recognition, presenting a variety of advantages over human‐based surveys. However, estimating abundance with such data collection techniques remains challenging because most abundance models require data that are difficult for low‐cost monoaural ARUs to gather (e.g., counts of individuals, distance to individuals), especially when using the output of automated species recognition. Statistical models that do not require counting or measuring distances to target individuals in combination with low‐cost ARUs provide a promising way of obtaining abundance estimates for large‐scale wildlife monitoring projects but remain untested. We present a case study using avian field data collected in the forests of Pennsylvania during the spring of 2020 and 2021 using both traditional point counts and passive acoustic monitoring at the same locations. We tested the ability of the Royle–Nichols and time‐to‐detection models to estimate the abundance of two species from detection histories generated by applying a machine‐learning classifier to ARU‐gathered data. We compared abundance estimates from these models with estimates from the same models fit using point‐count data and to two additional models appropriate for point counts, the N‐mixture model and distance models. We found that the Royle–Nichols and time‐to‐detection models can be used with ARU data to produce abundance estimates similar to those generated by a point‐count‐based study but with greater precision. ARU‐based models produced confidence or credible intervals that were on average 31.9% (±11.9 SE) smaller than their point‐count counterpart. Our findings were consistent across two species with differing relative abundance and habitat use patterns. The higher precision of models fit using ARU data is likely due to higher cumulative detection probability, which itself may be the result of greater survey effort using ARUs and machine‐learning classifiers to sample significantly more time for focal species at any given point. Our results provide preliminary support for the use of ARUs in abundance‐based study applications, and thus may afford researchers a better understanding of habitat quality and population trends, while allowing them to make more informed conservation recommendations and actions.

show abstract

Deep learning in marine bioacoustics: a benchmark for baleen whale detection

Schall,

Kaya,

Debusschere

et al. 2024

Remote Sens Ecol Conserv

View full text Add to dashboard Cite

Passive acoustic monitoring (PAM) is commonly used to obtain year‐round continuous data on marine soundscapes harboring valuable information on species distributions or ecosystem dynamics. This continuously increasing amount of data requires highly efficient automated analysis techniques in order to exploit the full potential of the available data. Here, we propose a benchmark, which consists of a public dataset, a well‐defined task and evaluation procedure to develop and test automated analysis techniques. This benchmark focuses on the special case of detecting animal vocalizations in a real‐world dataset from the marine realm. We believe that such a benchmark is necessary to monitor the progress in the development of new detection algorithms in the field of marine bioacoustics. We ultimately use the proposed benchmark to test three detection approaches, namely ANIMAL‐SPOT, Koogu and a simple custom sequential convolutional neural network (CNN), and report performances. We report the performance of the three detection approaches in a blocked cross‐validation fashion with 11 site‐year blocks for a multi‐species detection scenario in a large marine passive acoustic dataset. Performance was measured with three simple metrics (i.e., true classification rate, noise misclassification rate and call misclassification rate) and one combined fitness metric, which allocates more weight to the minimization of false positives created by noise. Overall, ANIMAL‐SPOT performed the best with an average fitness metric of 0.6, followed by the custom CNN with an average fitness metric of 0.57 and finally Koogu with an average fitness metric of 0.42. The presented benchmark is an important step to advance in the automatic processing of the continuously growing amount of PAM data that are collected throughout the world's oceans. To ultimately achieve usability of developed algorithms, the focus of future work should be laid on the reduction of the false positives created by noise.

show abstract

OpenSoundscape: An open‐source bioacoustics analysis package for Python

Cited by 15 publications

References 44 publications

Implications of the scale of detection for inferring co‐occurrence patterns from paired camera traps and acoustic recorders

Implications of the scale of detection for inferring co‐occurrence patterns from paired camera traps and acoustic recorders

Performance of unmarked abundance models with data from machine‐learning classification of passive acoustic recordings

Deep learning in marine bioacoustics: a benchmark for baleen whale detection

Contact Info

Product

Resources

About