Machine learning based animal emotion classification using audio signals

This paper presents the machine learning approach to the automated classification of a dog's emotional state based on the processing and recognition of audio signals. It offers helpful information for improving human-machine interfaces and developing more precise tools for classifying emotions from acoustic data. The presented model demonstrates an overall accuracy value above 70% for audio signals recorded for one dog.

@article{slobodian2025_2503.18138,
  title={ Machine learning based animal emotion classification using audio signals },
  author={ Mariia Slobodian and Mykola Kozlenko },
  journal={arXiv preprint arXiv:2503.18138},
  year={ 2025 }
}