Page 12 - 《应用声学》2025年第3期

546 2025 年 5 月

向还包括开发更鲁棒的模型结构，以及探索实际场 2024, 40(7): 1197–1207.
景中的高效部署。 [13] Kuang K L, Yang F R, Li J F, et al. Three-stage hy-
brid neural beamformer for multi-channel speech enhance-
ment[J]. The Journal of the Acoustical Society of America,
参 考 文 献 2023, 153(6): 3378.
[14] Yoon S, Park S, Yoo S. Two-stage adaptive noise reduction
system for broadcasting multicopters[C]//IEEE Interna-
[1] Martinez-Carranza J, Rascon C. A review on auditory
tional Conference on Consumer Electronics. IEEE, 2016:
perception for unmanned aerial vehicles[J]. Sensors, 2020,
219–222.
20(24): 7276.
[15] Wang L, Cavallaro A. Microphone-array ego-noise reduc-
[2] Deleforge A. Drone audition for search and rescue:
tion algorithms for auditory micro aerial vehicles[J]. IEEE
Datasets and challenges[C]//QUIET DRONES Interna-
Sensors Journal, 2017, 17(8): 2447–2455.
tional Symposium on UAV/UAS Noise. HAL, 2020.
[16] Wang L, Cavallaro A. Acoustic sensing from a multi-rotor
[3] Strauss M, Mordel P, Miguet V, et al. DREGON: Dataset
drone[J]. IEEE Sensors Journal, 2018, 18(11): 4570–4582.
and methods for UAV-embedded sound source localiza-
[17] Mukhutdinov D, Alex A, Cavallaro A, et al. Deep learn-
tion[C]//IEEE/RSJ International Conference on Intelli-
ing models for single-channel speech enhancement on
gent Robots and Systems. IEEE, 2018: 1–8.
drones[J]. IEEE Access, 2023, 11: 22993–23007.
[4] Fernandes R P, Santos E C. A first approach to sig-
[18] Tan Z W, Nguyen A H T, Khong A W H. An efficient
nal enhancement for quadcopters using piezoelectric sen-
dilated convolutional neural network for UAV noise re-
sors[C]//International Conference on Transformative Sci-
duction at low input SNR[C]//2019 Asia-Pacific Signal
ence and Engineering, Business and Social Innovation,
and Information Processing Association Annual Summit
2015: 536–541.
and Conference. IEEE, 2019: 1885–1892.
[5] Yen B, Li Y, Hioka Y. Rotor noise-aware noise covari-
[19] Marmaroli P, Falourd X, Lissek H. A UAV motor de-
ance matrix estimation for unmanned aerial vehicle audi-
noising technique to improve localization of surrounding
tion[J]. IEEE/ACM Transactions on Audio, Speech, and
noisy aircrafts: Proof of concept for anti-collision sys-
Language Processing, 2023, 31: 2491–2506.
tems[C]//Acoustics 2012, 2012.
[6] Parchami M, Zhu W P, Champagne B, et al. Recent devel-
[20] Wang L, Cavallaro A. Ear in the sky: Ego-noise reduction
opments in speech enhancement in the short-time Fourier
for auditory micro aerial vehicles[C]//13th IEEE Interna-
transform domain[J]. IEEE Circuits and Systems Maga-
tional Conference on Advanced Video and Signal Based
zine, 2016, 16(3): 45–77.
Surveillance (AVSS). IEEE, 2016: 152–158.
[7] Priyanka S S. A review on adaptive beamforming tech-
niques for speech enhancement[C]//Innovations in Power [21] Sun Y M, Zhang F M, Liu Y, et al. Acoustic event
and Advanced Computing Technologies. IEEE, 2017: 1–6. detection for drone search and rescue system based on
bi-directional long and short-term memory beamforming
[8] Zheng C S, Zhang H Y, Liu W Z, et al. Sixty years of
method to remove rotor noise[J]. Digital Signal Process-
frequency-domain monaural speech enhancement: From
ing, 2025, 157: 104881.
traditional to deep learning methods[J]. Trends in Hear-
ing, 2023, 27: 23312165231209913. [22] Bin Abdul Qayyum A, Naimul Hassan K M, Anika A,
[9] Wang Z Q, Wichern G, Watanabe S, et al. STFT-domain et al. DOANet: A deep dilated convolutional neural net-
neural speech enhancement with very low algorithmic la- work approach for search and rescue with drone-embedded
tency[J]. IEEE/ACM Transactions on Audio, Speech, and sound source localization[J]. EURASIP Journal on Audio,
Language Processing, 2022, 31: 397–410. Speech, and Music Processing, 2020, 2020(1): 16.
[10] Hu Y X, Liu Y, Lyu S B, et al. DCCRN: Deep [23] Jaiswal R. Speech activity detection under adverse noisy
complex convolution recurrent network for phase-aware conditions at low SNRs[C]//2021 6th International Con-
speech enhancement[C]//Interspeech 2020. ISCA, 2020: ference on Communication and Electronics Systems.
2472–2476. IEEE, 2021: 97–101.
[11] 王玫, 李江和, 宋浠瑜, 等. 基于轻量级卷积门控循环神经网 [24] Nakadai K, Kumon M, Okuno H G, et al. Development
络的语声增强方法 [J]. 应用声学, 2023, 42(3): 652–658. of microphone-array-embedded UAV for search and rescue
Wang Mei, Li Jianghe, Song Xiyu, et al. Speech enhance- task[C]//2017 IEEE/RSJ International Conference on In-
ment method based on lightweight convolution gated re- telligent Robots and Systems. IEEE, 2017: 5985–5990.
current neural network[J]. Journal of Applied Acoustics, [25] Wang L, Cavallaro A. A blind source separation frame-
2023, 42(3): 652–658. work for ego-noise reduction on multi-rotor drones[J].
[12] 雷菁, 王劲夫, 杨飞然, 等. 神经网络辅助估计先验语音存在概 IEEE/ACM Transactions on Audio, Speech, and Lan-
率的多通道降噪方法 [J]. 信号处理, 2024, 40(7): 1197–1207. guage Processing, 2020, 28: 2523–2537.
Lei Jing, Wang Jinfu, Yang Feiran, et al. NN-supported [26] Choi J, Chang J H. Convolutional neural network-based
a priori speech presence probability estimation for multi- direction-of-arrival estimation using stereo microphones
channel noise reduction[J]. Journal of Signal Processing, for drone[C]//2020 International Conference on Electron-