Page 61 - 《应用声学》2025年第3期

第 44 卷 第 3 期 杨雪同等： 基于注意力的双层级并行声学场景分类方法 595

Processing (ICASSP). IEEE, 2019: 56–60. [20] Ren S, He K, Girshick R, et al. Faster R-CNN: To-
[14] Gou J, Yu B, Maybank S J, et al. Knowledge distillation: wards real-time object detection with region proposal net-
A survey[J]. International Journal of Computer Vision, works[J]. IEEE Transactions on Pattern Analysis and Ma-
2021, 129: 1789–1819. chine Intelligence, 2017, 39(6): 1137–1149.
[15] Wang K, Gao X, Zhao Y, et al. Pay attention to features, [21] Woo S, Park J, Lee J Y, et al. Cbam: Convolutional block
transfer learn faster CNNs[C]// International Conference attention module[C]// European Conference on Com-
on Learning Representations, 2019. puter Vision (ECCV), 2018: 3–19.
[16] Liu M, Wang W, Li Y. The system for acoustic scene clas- [22] Hu J, Shen L, Sun G. Squeeze-and-excitation net-
sification using resnet[C]// IEEE AASP Challenge on De- works[C]// IEEE/CVF Conference on Computer Vision
tection and Classification of Acoustic Scenes and Events and Pattern Recognition (CVPR), 2018: 7132–7141.
(DCASE), 2019. [23] Plata M. Deep neural networks with supported clus-
[17] Salamon J, Bello J P. Deep convolutional neural networks ters preclassification procedure for acoustic scene recogni-
and data augmentation for environmental sound classi- tion[C]// IEEE AASP Challenge on Detection and Clas-
fication[J]. IEEE Signal Processing Letters, 2017, 24(3): sification of Acoustic Scenes and Events (DCASE), 2019.
279–283. [24] Ma S, Liu W. Acoustic scene classification based on bin-
[18] Hershey S, Chaudhuri S, Ellis D P W, et al. CNN ar- aural deep scattering spectra with neural network[C]//
chitectures for large-scale audio classification[C]// 2017 IEEE AASP Challenge on Detection and Classification of
IEEE International Conference on Acoustics, Speech and Acoustic Scenes and Events (DCASE), 2019.
Signal Processing (ICASSP). IEEE, 2017: 131–135. [25] Hou Y, Kang B, Mitchell A, et al. Cooperative scene-event
[19] Vaswani A, Shazeer N, Parmar N, et al. Attention is all modelling for acoustic scene classification[J]. IEEE/ACM
you need[C]// Advances in Neural Information Processing Transactions on Audio, Speech, and Language Processing,
Systems, 2017: 5998–6008. 2023, 32: 68–82.