Multi-level graph learning for audio event classification and human-perceived annoyance rating prediction
- Author
- Yuanbo Hou (UGent) , Qiaoqiao Ren (UGent) , Siyang Song, Yuxin Song, Wenwu Wang and Dick Botteldooren (UGent)
- Organization
- Project
- Abstract
- WHO’s report on environmental noise estimates that 22 M people suffer from chronic annoyance related to noise caused by audio events (AEs) from various sources. Annoyance may lead to health issues and adverse effects on metabolic and cognitive systems. In cities, monitoring noise levels does not provide insights into noticeable AEs, let alone their relations to annoyance. To create annoyance-related monitoring, this paper proposes a graph-based model to identify AEs in a sound-scape, and explore relations between diverse AEs and human-perceived annoyance rating (AR). Specifically, this paper proposes a lightweight multi-level graph learning (MLGL) based on local and global semantic graphs to simultaneously perform audio event classification (AEC) and human annoyance rating prediction (ARP). Experiments show that: 1) MLGL with 4.1 M parameters improves AEC and ARP results by using semantic node information in local and global context-aware graphs; 2) MLGL captures relations between coarse-and fine-grained AEs and AR well; 3) Statistical analysis of MLGL results shows that some AEs from different sources significantly correlate with AR, which is consistent with previous research on human perception of these sound sources.
Downloads
-
(...).pdf
- full text (Published version)
- |
- UGent only
- |
- |
- 13.19 MB
-
ACUS 717a.pdf
- full text (Accepted manuscript)
- |
- open access
- |
- |
- 12.28 MB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-01JHJNQDY925TR2H4VXKHVJ91M
- MLA
- Hou, Yuanbo, et al. “Multi-Level Graph Learning for Audio Event Classification and Human-Perceived Annoyance Rating Prediction.” ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), IEEE, 2024, pp. 716–20, doi:10.1109/ICASSP48485.2024.10446633.
- APA
- Hou, Y., Ren, Q., Song, S., Song, Y., Wang, W., & Botteldooren, D. (2024). Multi-level graph learning for audio event classification and human-perceived annoyance rating prediction. ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 716–720. https://doi.org/10.1109/ICASSP48485.2024.10446633
- Chicago author-date
- Hou, Yuanbo, Qiaoqiao Ren, Siyang Song, Yuxin Song, Wenwu Wang, and Dick Botteldooren. 2024. “Multi-Level Graph Learning for Audio Event Classification and Human-Perceived Annoyance Rating Prediction.” In ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 716–20. IEEE. https://doi.org/10.1109/ICASSP48485.2024.10446633.
- Chicago author-date (all authors)
- Hou, Yuanbo, Qiaoqiao Ren, Siyang Song, Yuxin Song, Wenwu Wang, and Dick Botteldooren. 2024. “Multi-Level Graph Learning for Audio Event Classification and Human-Perceived Annoyance Rating Prediction.” In ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 716–720. IEEE. doi:10.1109/ICASSP48485.2024.10446633.
- Vancouver
- 1.Hou Y, Ren Q, Song S, Song Y, Wang W, Botteldooren D. Multi-level graph learning for audio event classification and human-perceived annoyance rating prediction. In: ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE; 2024. p. 716–20.
- IEEE
- [1]Y. Hou, Q. Ren, S. Song, Y. Song, W. Wang, and D. Botteldooren, “Multi-level graph learning for audio event classification and human-perceived annoyance rating prediction,” in ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Seoul, Korea, 2024, pp. 716–720.
@inproceedings{01JHJNQDY925TR2H4VXKHVJ91M,
abstract = {{WHO’s report on environmental noise estimates that 22 M people suffer from chronic annoyance related to noise caused by audio events (AEs) from various sources. Annoyance may lead to health issues and adverse effects on metabolic and cognitive systems. In cities, monitoring noise levels does not provide insights into noticeable AEs, let alone their relations to annoyance. To create annoyance-related monitoring, this paper proposes a graph-based model to identify AEs in a sound-scape, and explore relations between diverse AEs and human-perceived annoyance rating (AR). Specifically, this paper proposes a lightweight multi-level graph learning (MLGL) based on local and global semantic graphs to simultaneously perform audio event classification (AEC) and human annoyance rating prediction (ARP). Experiments show that: 1) MLGL with 4.1 M parameters improves AEC and ARP results by using semantic node information in local and global context-aware graphs; 2) MLGL captures relations between coarse-and fine-grained AEs and AR well; 3) Statistical analysis of MLGL results shows that some AEs from different sources significantly correlate with AR, which is consistent with previous research on human perception of these sound sources.}},
author = {{Hou, Yuanbo and Ren, Qiaoqiao and Song, Siyang and Song, Yuxin and Wang, Wenwu and Botteldooren, Dick}},
booktitle = {{ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}},
isbn = {{9798350344868}},
issn = {{1520-6149}},
language = {{eng}},
location = {{Seoul, Korea}},
pages = {{716--720}},
publisher = {{IEEE}},
title = {{Multi-level graph learning for audio event classification and human-perceived annoyance rating prediction}},
url = {{http://doi.org/10.1109/ICASSP48485.2024.10446633}},
year = {{2024}},
}
- Altmetric
- View in Altmetric