Species-independent analysis and identification of emotional animal vocalizations

Abstract Animal vocalizations can differ depending on the context in which they are produced and serve as an instant indicator of an animal’s emotional state. Interestingly, from an evolutional perspective, it should be possible to directly compare different species using the same set of acoustic ma...

Full description

Saved in:
Bibliographic Details
Main Author: Stavros Ntalampiras
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-14323-2
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849333358025244672
author Stavros Ntalampiras
author_facet Stavros Ntalampiras
author_sort Stavros Ntalampiras
collection DOAJ
description Abstract Animal vocalizations can differ depending on the context in which they are produced and serve as an instant indicator of an animal’s emotional state. Interestingly, from an evolutional perspective, it should be possible to directly compare different species using the same set of acoustic markers. This paper proposes a deep neural network architecture for analysing and recognizing vocalizations representing positive and negative emotional states. Understanding these vocalizations is critical for advancing animal health and welfare, a subject of growing importance due to its ethical, environmental, economic, and public health implications. To this end, a framework assessing the relationships between vocalizations was developed. Towards keeping all potentially relevant audio content, the constructed framework operates on log-Mel spectrograms. Similarities/dissimilarities are learned by a suitably designed Siamese Neural Network composed of convolutional layers. The formed latent space is appropriately clustered to identify the support set facilitating the emotion classification task. We employed a publicly available dataset and followed a thorough experimental protocol. The efficacy of such a scheme is shown after extensive experiments considering both classification and support set selection. Last but not least, by elaborating collectively the network’s activations when processing positive and negative vocalizations, important differences in the time-frequency plane are evidenced across emotions and species, assisting their understanding from animal scientists.
format Article
id doaj-art-c468abd87b7b417bbfca0784cfc727a3
institution Kabale University
issn 2045-2322
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-c468abd87b7b417bbfca0784cfc727a32025-08-20T03:45:53ZengNature PortfolioScientific Reports2045-23222025-08-0115111010.1038/s41598-025-14323-2Species-independent analysis and identification of emotional animal vocalizationsStavros Ntalampiras0Department of Computer Science, University of MilanAbstract Animal vocalizations can differ depending on the context in which they are produced and serve as an instant indicator of an animal’s emotional state. Interestingly, from an evolutional perspective, it should be possible to directly compare different species using the same set of acoustic markers. This paper proposes a deep neural network architecture for analysing and recognizing vocalizations representing positive and negative emotional states. Understanding these vocalizations is critical for advancing animal health and welfare, a subject of growing importance due to its ethical, environmental, economic, and public health implications. To this end, a framework assessing the relationships between vocalizations was developed. Towards keeping all potentially relevant audio content, the constructed framework operates on log-Mel spectrograms. Similarities/dissimilarities are learned by a suitably designed Siamese Neural Network composed of convolutional layers. The formed latent space is appropriately clustered to identify the support set facilitating the emotion classification task. We employed a publicly available dataset and followed a thorough experimental protocol. The efficacy of such a scheme is shown after extensive experiments considering both classification and support set selection. Last but not least, by elaborating collectively the network’s activations when processing positive and negative vocalizations, important differences in the time-frequency plane are evidenced across emotions and species, assisting their understanding from animal scientists.https://doi.org/10.1038/s41598-025-14323-2Animal health and welfareAudio pattern recognitionSpectral clusteringLatent representation
spellingShingle Stavros Ntalampiras
Species-independent analysis and identification of emotional animal vocalizations
Scientific Reports
Animal health and welfare
Audio pattern recognition
Spectral clustering
Latent representation
title Species-independent analysis and identification of emotional animal vocalizations
title_full Species-independent analysis and identification of emotional animal vocalizations
title_fullStr Species-independent analysis and identification of emotional animal vocalizations
title_full_unstemmed Species-independent analysis and identification of emotional animal vocalizations
title_short Species-independent analysis and identification of emotional animal vocalizations
title_sort species independent analysis and identification of emotional animal vocalizations
topic Animal health and welfare
Audio pattern recognition
Spectral clustering
Latent representation
url https://doi.org/10.1038/s41598-025-14323-2
work_keys_str_mv AT stavrosntalampiras speciesindependentanalysisandidentificationofemotionalanimalvocalizations