ESC-50

The dataset used for training the CNN in cough detection is composed of various modified audio clips gathered from open-source online sources. Each of these audio files originally contained at least one cough event and are cropped to a length of five seconds with the full cough contained at some arbitrary point within.

Data and Resources

Cite this as

K. J. Piczak (2024). Dataset: ESC-50. https://doi.org/10.57702/pa4431n8

DOI retrieved: December 2, 2024

Additional Info

Field Value
Created December 2, 2024
Last update December 2, 2024
Defined In https://doi.org/10.48550/arXiv.2309.12259
Citation
  • https://doi.org/10.1145/3491102.3502102
  • https://doi.org/10.48550/arXiv.2407.12707
  • https://doi.org/10.48550/arXiv.2103.02420
  • https://doi.org/10.48550/arXiv.2203.06059
  • https://doi.org/10.48550/arXiv.2202.00874
  • https://doi.org/10.48550/arXiv.2207.07429
  • https://doi.org/10.1109/EHB50910.2020.9280115
  • https://doi.org/10.48550/arXiv.1711.10282
  • https://doi.org/10.48550/arXiv.2311.18399
  • https://doi.org/10.48550/arXiv.2305.07132
Author K. J. Piczak
Homepage https://www.kaggle.com/c/esc-50