A Whisker of Truth: A Multimodal Interdisciplinary Machine Learning Approach to Vocal, Visual, and Tactile Signals in the Domestic Cat
(2025) p.1-8- Abstract
- We propose a multimodal deep learning framework for automated analysis of
cat–human communication, integrating acoustic, visual, and tactile signals through
transformer-based fusion. Using the largest expert-annotated dataset of its kind
and interdisciplinary collaboration, we combine semi-supervised learning with
ethological and phonetic expertise to detect subtle behavioural and phonetic cues,
enable early welfare assessment, and establish species-generalisable methods. - Abstract (Swedish)
- We propose a multimodal deep learning framework for automated analysis of
cat–human communication, integrating acoustic, visual, and tactile signals through
transformer-based fusion. Using the largest expert-annotated dataset of its kind
and interdisciplinary collaboration, we combine semi-supervised learning with
ethological and phonetic expertise to detect subtle behavioural and phonetic cues,
enable early welfare assessment, and establish species-generalisable methods.
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/18106a7a-f7a0-4be1-ad8d-fa3bbaca225d
- author
- van Toor, Astrid
; Schötz, Susanne
LU
and Hirsch, Elin
LU
- organization
- publishing date
- 2025
- type
- Contribution to conference
- publication status
- published
- subject
- pages
- 1 - 8
- project
- Melody in human–cat communication
- Interspecific Communication ASG, Pufendorf IAS
- Cat-human communication: vocal, visual and tactile signals
- language
- English
- LU publication?
- yes
- id
- 18106a7a-f7a0-4be1-ad8d-fa3bbaca225d
- alternative location
- https://openreview.net/attachment?id=HIkzPhD0Zo&name=pdf
- date added to LUP
- 2026-01-13 10:14:46
- date last changed
- 2026-01-14 07:36:53
@misc{18106a7a-f7a0-4be1-ad8d-fa3bbaca225d,
abstract = {{We propose a multimodal deep learning framework for automated analysis of<br/>cat–human communication, integrating acoustic, visual, and tactile signals through<br/>transformer-based fusion. Using the largest expert-annotated dataset of its kind<br/>and interdisciplinary collaboration, we combine semi-supervised learning with<br/>ethological and phonetic expertise to detect subtle behavioural and phonetic cues,<br/>enable early welfare assessment, and establish species-generalisable methods.}},
author = {{van Toor, Astrid and Schötz, Susanne and Hirsch, Elin}},
language = {{eng}},
pages = {{1--8}},
title = {{A Whisker of Truth: A Multimodal Interdisciplinary Machine Learning Approach to Vocal, Visual, and Tactile Signals in the Domestic Cat}},
url = {{https://lup.lub.lu.se/search/files/239402649/6_A_Whisker_of_Truth_A_Multimo.pdf}},
year = {{2025}},
}