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

Romero‐Mujalli, Daniel; Bergmann, Tjard; Zimmermann, A; Scheumann, Marina

doi:10.1038/s41598-021-03941-1

Cited by 10 publications

(12 citation statements)

References 64 publications

(77 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is consistent with the findings that each species may vary in the detection based on specific signatures (Scott Brandes 2008). In this case, the two types of vocalisations of a species are detected differently (Romero-Mujalli et al 2021). Our assessment of the use of bioacoustics as a survey method included an assessment of the detection distance of the bird’s call in its habitat.…”

Section: Discussionmentioning

confidence: 99%

Playback-aided surveys and acoustic monitoring in the detection of the Endangered Forest OwletAthene blewitti

Rajan,

Neema,

Trivedi

et al. 2024

Preprint

View full text Add to dashboard Cite

Long-term monitoring of populations of rare, endangered species is often challenging. Both the availability of baseline historical datasets and appropriate methods for long-term monitoring are often limited. Anthropogenic climate change and landscape change can impact species distributions significantly, sometimes resulting in a distributional shift of species and, in some cases, driving species to extinction. The Forest Owlet is an endangered bird that was considered extinct but was rediscovered after 113 years in 1997. Since its rediscovery, followed by the description of its calls, there have been regular sightings of the species from newer locations, leading to its down-listing in the IUCN Red List from Critically Endangered to Endangered. One area of interest is the Dang region in Gujarat, India, where there have been no historical records despite previous ornithological studies, but there are several recent records. Through field surveys, we repurpose data from the last three decades (1990, 2000, and 2019) to explore if this bird currently occurs in previous study sites where it was not recorded. The period has seen the rise of acoustic data, and we assess if new survey techniques using playback of its call could increase its detection. Additionally, we examined any changes in landscape and climate in this region across the same period. We also developed an acoustic detection framework for detecting the Forest Owlet from co-occurring sympatric owlets using Automated Recording Units (ARU) and sound-analysis software. We assessed appropriate detection distances from vocalizing birds on the field to design a spacing grid for future surveys of the species. We could locate the Forest Owlet from the resurveys at locations where previous studies three decades ago had not. We also find a significant change in the landscape - loss of native forests and increased agriculture, along with a significant change in climatic variables - mean maximum temperature and mean rainfall. Although the detection of the Forest Owlet is higher when accompanied by playback of its call, there is considerable variation across the landscape. Our acoustic detector comparison led us to a detection strategy for long-term monitoring; different approaches for songs and calls, and an effective detection distance of 300m in its habitat. Although the species responds to climate and habitat change, our study cannot determine the cause of the increased reports of this endangered species. All possibilities remain; the increase in the recent records of the species could be from variable detection or changes in climate and land use. However, we do find increased detections with newer survey techniques with bioacoustics, and we recommend they be used with care for future baseline studies that are urgently required.

show abstract

Section: Discussionmentioning

confidence: 99%

Playback-aided surveys and acoustic monitoring in the detection of the Endangered Forest OwletAthene blewitti

Rajan,

Neema,

Trivedi

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Higher endurance broadband recorders are now available, opening the potential to use new automated pattern recognition techniques to routinely extract distinct animal vocalizations from continuous long-term recordings [ 63 , 64 ]. These, in turn, could be used to explore other aspects of foraging theory, including investigating how patterns of bray call production within encounters affect decisions over when to leave prey patches [ 51 ].…”

Section: Discussionmentioning

confidence: 99%

Variation in foraging activity influences area-restricted search behaviour by bottlenose dolphins

et al. 2023

View full text Add to dashboard Cite

Area-restricted search (ARS) behaviour is commonly used to characterize spatio-temporal variation in foraging activity of predators, but evidence of the drivers underlying this behaviour in marine systems is sparse. Advances in underwater sound recording techniques and automated processing of acoustic data now provide opportunities to investigate these questions where species use different vocalizations when encountering prey. Here, we used passive acoustics to investigate drivers of ARS behaviour in a population of dolphins and determined if residency in key foraging areas increased following encounters with prey. Analyses were based on two independent proxies of foraging: echolocation buzzes (widely used as foraging proxies) and bray calls (vocalizations linked to salmon predation attempts). Echolocation buzzes were extracted from echolocation data loggers and bray calls from broadband recordings by a convolutional neural network. We found a strong positive relationship between the duration of encounters and the frequency of both foraging proxies, supporting the theory that bottlenose dolphins engage in ARS behaviour in response to higher prey encounter rates. This study provides empirical evidence for one driver of ARS behaviour and demonstrates the potential for applying passive acoustic monitoring in combination with deep learning-based techniques to investigate the behaviour of vocal animals.

show abstract

“…The human auditory range is limited, and mammals frequently produce and perceive sound at frequencies beyond human auditory abilities (Heffner and Heffner, 2018). Both infrasound and ultrasound are used by mammals in terrestrial and aquatic habitats (Martin et al, 2017) and detection of these vocalizations require specialized bioacoustics monitoring equipment and this fact may help to explain the paucity of data (Ladich and Winkler, 2017;Romero-Mujalli et al, 2021).…”

Section: Discussion and Future Directionsmentioning

confidence: 99%

Perspectives on Antiphonal Calling, Duetting and Counter-Singing in Non-primate Mammals: An Overview With Notes on the Coordinated Vocalizations of Bamboo Rats (Dactylomys spp., Rodentia: Echimyidae)

Vanderhoff

Hoverud

2022

Front. Ecol. Evol.

View full text Add to dashboard Cite

Temporally coordinated interactive vocalizations are important means of communication between individuals in various animal taxa. In mammals, interactive calling and singing can be highly synchronized to create either overlapping or antiphonal duets while in others, competitors antagonistically vocalize, engaging in counter-singing. Among non-primate mammals these vocalizations are considered rare and poorly understood. We provide an overview of antiphonal calling, duetting and counter-singing in non-primate mammals. Many of these coordinated vocalizations play a role in social interactions and allow mammals to convey information to other members of the social unit in visually inaccessible environments. South American Bamboo rats Dactylomys spp. are arboreal bamboo specialists found in dense bamboo thickets in Bolivia, Peru, Ecuador, Brazil and Colombia. These nocturnal rodents are rarely seen but can be easily heard because of their loud and distinctive staccato vocalizations. We provide some evidence that Bamboo rats engage in duetting, and as such they provide another case of a mammalian species, in which to investigate temporally coordinated interactive singing. We urge researchers to work toward common definitions of temporally coordinated vocalizations and to search for more mammals that utilize such vocalizations.

show abstract

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

Cited by 10 publications

References 64 publications

Playback-aided surveys and acoustic monitoring in the detection of the Endangered Forest OwletAthene blewitti

Playback-aided surveys and acoustic monitoring in the detection of the Endangered Forest OwletAthene blewitti

Variation in foraging activity influences area-restricted search behaviour by bottlenose dolphins

Perspectives on Antiphonal Calling, Duetting and Counter-Singing in Non-primate Mammals: An Overview With Notes on the Coordinated Vocalizations of Bamboo Rats (Dactylomys spp., Rodentia: Echimyidae)

Contact Info

Product

Resources

About