Exploring annoyance in a soundscape context by joint prediction of sound source and annoyance
- Author
- Yuanbo Hou (UGent) , Andrew Mitchell, Qiaoqiao Ren (UGent) , Francesco Aletta, Jian Kang and Dick Botteldooren (UGent)
- Organization
- Project
- Abstract
- Soundscape, the sonic environment as perceived and understood by people, is a conglomerate of different sounds. It has been established that its appraisal by instantaneous annoyance is not solely determined by its calculated loudness, but also by recognised sounds. Hence, most previous research on annoyance has focused on single-source environments. Audio analytics aims at detecting and classifying sound sources, but does not explore human perception of these. This paper proposes a dual-input model to simultaneously perform sound source classification (SSC) and human annoyance rating prediction (ARP). The model takes mel features and root-mean-square value (rms) features as input, and uses convolutional blocks to extract high-level acoustic features. These are used to predict sound source classes and to estimate the human annoyance rating for the whole fragment. Experiments on the DeLTA dataset show that: 1) models using mel features and rms features outperform models using only one of them; 2) The proposed model achieves a SSC accuracy of 90.06%, and an ARP (scale 1 to 10) root mean square error of 1.05.
Downloads
-
ACUS 652.pdf
- full text (Published version)
- |
- open access
- |
- |
- 1.92 MB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-01HHHRMB1CFBX7E12YQXJ0G9P2
- MLA
- Hou, Yuanbo, et al. “Exploring Annoyance in a Soundscape Context by Joint Prediction of Sound Source and Annoyance.” Forum Acusticum 2023 : 10th Convention of the European Acoustics Association, Proceedings, 2023, pp. 1–5.
- APA
- Hou, Y., Mitchell, A., Ren, Q., Aletta, F., Kang, J., & Botteldooren, D. (2023). Exploring annoyance in a soundscape context by joint prediction of sound source and annoyance. Forum Acusticum 2023 : 10th Convention of the European Acoustics Association, Proceedings, 1–5.
- Chicago author-date
- Hou, Yuanbo, Andrew Mitchell, Qiaoqiao Ren, Francesco Aletta, Jian Kang, and Dick Botteldooren. 2023. “Exploring Annoyance in a Soundscape Context by Joint Prediction of Sound Source and Annoyance.” In Forum Acusticum 2023 : 10th Convention of the European Acoustics Association, Proceedings, 1–5.
- Chicago author-date (all authors)
- Hou, Yuanbo, Andrew Mitchell, Qiaoqiao Ren, Francesco Aletta, Jian Kang, and Dick Botteldooren. 2023. “Exploring Annoyance in a Soundscape Context by Joint Prediction of Sound Source and Annoyance.” In Forum Acusticum 2023 : 10th Convention of the European Acoustics Association, Proceedings, 1–5.
- Vancouver
- 1.Hou Y, Mitchell A, Ren Q, Aletta F, Kang J, Botteldooren D. Exploring annoyance in a soundscape context by joint prediction of sound source and annoyance. In: Forum Acusticum 2023 : 10th Convention of the European Acoustics Association, Proceedings. 2023. p. 1–5.
- IEEE
- [1]Y. Hou, A. Mitchell, Q. Ren, F. Aletta, J. Kang, and D. Botteldooren, “Exploring annoyance in a soundscape context by joint prediction of sound source and annoyance,” in Forum Acusticum 2023 : 10th Convention of the European Acoustics Association, Proceedings, Turin, Italy, 2023, pp. 1–5.
@inproceedings{01HHHRMB1CFBX7E12YQXJ0G9P2,
abstract = {{Soundscape, the sonic environment as perceived and understood by people, is a conglomerate of different sounds.
It has been established that its appraisal by instantaneous
annoyance is not solely determined by its calculated loudness, but also by recognised sounds. Hence, most previous
research on annoyance has focused on single-source environments. Audio analytics aims at detecting and classifying sound sources, but does not explore human perception
of these. This paper proposes a dual-input model to simultaneously perform sound source classification (SSC) and
human annoyance rating prediction (ARP). The model
takes mel features and root-mean-square value (rms) features as input, and uses convolutional blocks to extract
high-level acoustic features. These are used to predict
sound source classes and to estimate the human annoyance rating for the whole fragment. Experiments on the
DeLTA dataset show that: 1) models using mel features
and rms features outperform models using only one of
them; 2) The proposed model achieves a SSC accuracy
of 90.06%, and an ARP (scale 1 to 10) root mean square
error of 1.05.}},
author = {{Hou, Yuanbo and Mitchell, Andrew and Ren, Qiaoqiao and Aletta, Francesco and Kang, Jian and Botteldooren, Dick}},
booktitle = {{Forum Acusticum 2023 : 10th Convention of the European Acoustics Association, Proceedings}},
isbn = {{9788888942674}},
issn = {{2221-3767}},
language = {{eng}},
location = {{Turin, Italy}},
pages = {{1--5}},
title = {{Exploring annoyance in a soundscape context by joint prediction of sound source and annoyance}},
year = {{2023}},
}