Nguyen Duc Toan1, Quan Trong The2,*, Nguyen Minh Thang3
1Information Technology Institute, Vietnam Women's Academy, Hanoi 11528, Vietnam.
2Lab Blockchain, Faculty of Information Security, Posts and Telecommunications Institute of Technology (PTIT), Hanoi 12108, Vietnam.
3Faculty of Information Technology, Ho Chi Minh City University of Economics and Finance, Ho Chi Minh City 700000, Vietnam.
*Corresponding author:Quan Trong The
This work was sponsored by Posts and Telecommunications Institute of Technology (PTIT), Hanoi, Vietnam.
Abstract
Nowadays, microphone array (MA) beamforming has been popular due to its convenience of easy installation, the high directivity index, and the capability of steering the steerable beampattern to a certain sound location while suppressing surrounding noise, interference, and signals from other directions. MA beamformers are commonly integrated into various types of speech applications, such as teleconference systems, hearing aids, voice-controlled devices, surveillance equipment, smart phone. Generalized Sidelobe Canceller (GSC) beamformer based on the constrained criteria of preserving the original clean speech data while mitigating the background acoustic environment with minimum noise power. However, under realistic recording scenarios, due to the complex and annoying situation, the existence of non-directional noise, the different sensitivities of microphones, the sampling rate, and the movement of the talker, the GSC beamformer’s performance is often corrupted. In this paper, the author proposed a soft decision, which allows obtaining the exact speech presence probability to increase the GSC beamformer’s evaluation. The numerical has shown improvement in reducing the speech distortion, removing the residual noise component, and increasing the speech quality in terms of signal-to-noise (SNR) ratio.
References
[1] Fang R, et al. A frequency invariant broadband beamformer based on directional microphone array. In: 2022 14th International Conference on Signal Processing Systems (ICSPS). Jiangsu, China; 2022. p. 623-7.
doi: 10.1109/ICSPS58776.2022.00115
[2] Hossein MT, Hossain MS, Rashid M, Ahmed N. Robust microphone beamformer in the presence of multiple interfering signals. In: 2021 3rd International Conference on Electrical & Electronic Engineering (ICEEE). Rajshahi, Bangladesh; 2021. p. 173-6.
doi: 10.1109/ICEEE54059.2021.9718861
[3] Zhang F, Benesty J, Pan C, Chen J. A universal linear beamformer for microphone arrays. In: 2025 IEEE 15th International Conference on Signal Processing, Communications and Computing (ICSPCC). Hong Kong, Hong Kong; 2025. p. 1-5.
doi: 10.1109/ICSPCC66825.2025.11194607
[4] Lee C-H, Yang C, Shen Y, Jin H. Improved mask-based neural beamforming for multichannel speech enhancement by snapshot matching masking. In: ICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Rhodes Island, Greece; 2023. p. 1-5. doi: 10.1109/ICASSP49357.2023.10096213
[5] Zhang F, Pan C, Benesty J, Chen J. Directional gain based noise covariance matrix estimation for MVDR beamforming. In: ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Seoul, Korea, Republic of; 2024. p. 511-5.
doi: 10.1109/ICASSP48485.2024.10447393
[6] Zhang F, Pan C, Benesty J, Chen J. Simplified maximum SNR beamformers with spatial coherence matrix modeling. In: 2023 31st European Signal Processing Conference (EUSIPCO). Helsinki, Finland; 2023. p. 6-10.
doi: 10.23919/EUSIPCO58844.2023.10289966
[7] Zhang F, Pan C, Benesty J, Chen J. On the design and implementation of maximum SNR beamformers for scattered speech sources. In: 2023 IEEE International Conference on Signal Processing, Communications and Computing (ICSPCC). Zhengzhou, China; 2023. p. 1-6.
doi: 10.1109/ICSPCC59353.2023.10400316
[8] Liang J, Zeng Q. Improved spectral subtraction based on second-order differential array and phase spectrum compensation. In: 2023 3rd International Symposium on Computer Technology and Information Science (ISCTIS). Chengdu, China; 2023. p. 658-61.
doi: 10.1109/ISCTIS58954.2023.10212993
[9] Ito N, Ono N, Vincent E, Sagayama S. Designing the Wiener post-filter for diffuse noise suppression using imaginary parts of inter-channel cross-spectra. In: 2010 IEEE International Conference on Acoustics, Speech and Signal Processing. Dallas, TX, USA; 2010. p. 2818-21.
doi: 10.1109/ICASSP.2010.5496202
[10] Yang X, Huang G, Jin J, Chen J, Benesty J. Design and optimization of superdirective beamforming and post-filtering for speech enhancement. In: ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Hyderabad, India; 2025. p. 1-5.
doi: 10.1109/ICASSP49660.2025.10890758
[11] Kim SM. Hearing aid speech enhancement using phase difference-controlled dual-microphone generalized sidelobe canceller. IEEE Access. 2019;7:130663-71.
doi: 10.1109/ACCESS.2019.2940047
[12] SNRVAD [Internet]. Available from: https://labrosa.ee.columbia.edu/projects/snreval/