Multi-Feature Audio Fusion for Nonverbal Vocalization Classification

Siddhant Bikram Shah and Kristina T. Johnson

This is the code repository for our paper Multi-Feature Audio Fusion for Nonverbal Vocalization Classification published at ICASSP 2025.

Code

All experimental changes can be made through a single file: configs.py.

Dataset

The open-access ReCANVo dataset was used for this paper. The dataset can be downloaded from this link.

Citation

@inproceedings{shah2025multi,
  title={Multi-Feature Audio Fusion for Nonverbal Vocalization Classification},
  author={Shah, Siddhant Bikram and Johnson, Kristina T},
  booktitle={ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  pages={1--5},
  year={2025},
  organization={IEEE}
}

OR

S. B. Shah and K. T. Johnson, "Multi-Feature Audio Fusion for Nonverbal Vocalization Classification," ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Hyderabad, India, 2025, pp. 1-5, doi: 10.1109/ICASSP49660.2025.10889317.