Acoustic auto-encoders for biodiversity assessment

Rowe, Benjamin; Eichinski, Philip; Zhang, Jinglan; Roe, Paul

doi:10.1016/j.ecoinf.2021.101237

Cited by 12 publications

(14 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An audio segment duration of twelve seconds is unlike the standard practice in the field, which is typically less than five seconds [ 19 ]. Our choice of 12 s segments was influenced by the Short-Time Fourier Transform (STFT) parameters, which yield 515 bins in the frequency axis.…”

Section: Methodsmentioning

confidence: 99%

“…The majority of deep learning methodologies for ecoacoustics applications reported in the literature rely on supervised approaches, whereas end-to-end unsupervised frameworks have not been sufficiently studied. However, some approximations have emerged, e.g., Rowe et al [ 19 ] used autoencoders to characterize sound types and identify groups corresponding to species vocalizations. Nevertheless, although the methodology is unsupervised, it requires prior information about the species for validation.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Soundscape Characterization Using Autoencoders and Unsupervised Learning

Nieto-Mora,

Ferreira de Oliveira,

Sanchez-Giraldo

et al. 2024

Sensors

View full text Add to dashboard Cite

Passive acoustic monitoring (PAM) through acoustic recorder units (ARUs) shows promise in detecting early landscape changes linked to functional and structural patterns, including species richness, acoustic diversity, community interactions, and human-induced threats. However, current approaches primarily rely on supervised methods, which require prior knowledge of collected datasets. This reliance poses challenges due to the large volumes of ARU data. In this work, we propose a non-supervised framework using autoencoders to extract soundscape features. We applied this framework to a dataset from Colombian landscapes captured by 31 audiomoth recorders. Our method generates clusters based on autoencoder features and represents cluster information with prototype spectrograms using centroid features and the decoder part of the neural network. Our analysis provides valuable insights into the distribution and temporal patterns of various sound compositions within the study area. By utilizing autoencoders, we identify significant soundscape patterns characterized by recurring and intense sound types across multiple frequency ranges. This comprehensive understanding of the study area’s soundscape allows us to pinpoint crucial sound sources and gain deeper insights into its acoustic environment. Our results encourage further exploration of unsupervised algorithms in soundscape analysis as a promising alternative path for understanding and monitoring environmental changes.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Soundscape Characterization Using Autoencoders and Unsupervised Learning

Nieto-Mora,

Ferreira de Oliveira,

Sanchez-Giraldo

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…Rather than designing acoustic indices heuristically, deep learning algorithms learn a numerical soundscape description from a given audio set. These can be with pre‐trained models (Sethi et al, 2020) or use self‐supervised methods (Rowe et al, 2021). Learned representations are high dimensional, data‐driven descriptors and, unlike current acoustic indices, not based on human assumptions about links between soundscapes and ecology.…”

Section: Emerging Trends In Ecoacoustic Analysesmentioning

confidence: 99%

Using acoustic indices in ecology: Guidance on study design, analyses and interpretation

et al. 2023

View full text Add to dashboard Cite

The rise of passive acoustic monitoring and the rapid growth in large audio datasets is driving the development of analysis methods that allow ecological inferences to be drawn from acoustic data. Acoustic indices are currently one of the most widely applied tools in ecoacoustics. These numerical summaries of the sound energy contained in digital audio recordings are relatively straightforward and fast to calculate but can be challenging to interpret. Misapplication and misinterpretation have produced conflicting results and led some to question their value. To encourage better use of acoustic indices, we provide nine points of guidance to support good study design, analysis and interpretation. We offer practical recommendations for the use of acoustic indices in the study of both whole soundscapes and individual taxa and species, and point to emerging trends in ecoacoustic analysis. In particular, we highlight the critical importance of understanding the links between soundscape patterns and acoustic indices. Acoustic indices can offer insights into the state of organisms, populations, and ecosystems, complementing other ecological research techniques. Judicious selection, appropriate application and thorough interpretation of existing indices is vital to bolster robust developments in ecoacoustics for biodiversity monitoring, conservation and future research.

show abstract

“…An opportunity to avoid having to manually choose the right set of features and tune their settings is to use features learnt using deep representation learning (as opposed to handcrafted features like PAFs). Following this approach, auto-encoder artificial neural networks have been used by Goffinet et al [28] on mice and zebra finch vocalisations, by Bergler et al [29] to cluster orca calls, by Rowe et al [30] to cluster bird vocalisations by species, and by Tolkova et al [31] to discriminate between background noise and bird vocalisations.…”

Section: Vocalisations Feature Extraction and Clusteringmentioning

confidence: 99%

Deep audio embeddings for vocalisation clustering

et al. 2023

View full text Add to dashboard Cite

The study of non-human animals’ communication systems generally relies on the transcription of vocal sequences using a finite set of discrete units. This set is referred to as a vocal repertoire, which is specific to a species or a sub-group of a species. When conducted by human experts, the formal description of vocal repertoires can be laborious and/or biased. This motivates computerised assistance for this procedure, for which machine learning algorithms represent a good opportunity. Unsupervised clustering algorithms are suited for grouping close points together, provided a relevant representation. This paper therefore studies a new method for encoding vocalisations, allowing for automatic clustering to alleviate vocal repertoire characterisation. Borrowing from deep representation learning, we use a convolutional auto-encoder network to learn an abstract representation of vocalisations. We report on the quality of the learnt representation, as well as of state of the art methods, by quantifying their agreement with expert labelled vocalisation types from 8 datasets of other studies across 6 species (birds and marine mammals). With this benchmark, we demonstrate that using auto-encoders improves the relevance of vocalisation representation which serves repertoire characterisation using a very limited number of settings. We also publish a Python package for the bioacoustic community to train their own vocalisation auto-encoders or use a pretrained encoder to browse vocal repertoires and ease unit wise annotation.

show abstract

Acoustic auto-encoders for biodiversity assessment

Cited by 12 publications

References 14 publications

Soundscape Characterization Using Autoencoders and Unsupervised Learning

Soundscape Characterization Using Autoencoders and Unsupervised Learning

Using acoustic indices in ecology: Guidance on study design, analyses and interpretation

Deep audio embeddings for vocalisation clustering

Contact Info

Product

Resources

About