抑制多方向语音噪声的人工耳蜗语音增强算法

宫琴; 饶诚; 郑硕

doi:10.16511/j.cnki.qhdxxb.2019.21.022

PDF(1511 KB)

清华大学学报（自然科学版） ›› 2020, Vol. 60 ›› Issue (2) : 181-188. DOI: 10.16511/j.cnki.qhdxxb.2019.21.022

生物医学

抑制多方向语音噪声的人工耳蜗语音增强算法

宫琴^1,2, 饶诚¹, 郑硕¹

作者信息 +

Speech enhancement algorithm for cochlear implants to suppress multi-directional speech noise

GONG Qin^1,2, RAO Cheng¹, ZHENG Shuo¹

Author information +

文章历史 +

摘要

在复杂声音场景，尤其是多人对话交谈的情况中，人工耳蜗佩戴者的言语可懂度会显著下降。该文提出了一种基于近距离双麦克风的语音增强算法，通过提取采集信号的延迟参数，并根据该参数设置掩蔽矩阵，进而区分不同方位的声源，抑制竞争性语音噪声。同时还以数字信号处理器（digital signal processor，DSP）为核心，搭建调试平台，对算法进行了工程实现。实际测试表明，算法提升了前端信号的信噪比，鲁棒性强，并且算法单帧运行时间短，可以和常用的人工耳蜗言语编码策略相融合，满足人工耳蜗的实时计算需求。

Abstract

Cochlear implant users often cannot clearly hear sounds in complex sound environments, especially with competing talkers. This paper describes a speech enhancement algorithm based on closely-spaced microphones. The algorithm extracts a delay parameter and defines a mask matrix based on this parameter to distinguish sound sources having different orientations. The model uses a digital signal processor (DSP) to implement the algorithm. Tests indicate that the robust algorithm improves the signal-to-noise ratio. The algorithm has short single-frame run times and can be integrated with commonly used speech coding strategies, which are needed for real-time computing in cochlear implants.

导出引用

宫琴, 饶诚, 郑硕. 抑制多方向语音噪声的人工耳蜗语音增强算法[J]. 清华大学学报（自然科学版）. 2020, 60(2): 181-188 https://doi.org/10.16511/j.cnki.qhdxxb.2019.21.022

GONG Qin, RAO Cheng, ZHENG Shuo. Speech enhancement algorithm for cochlear implants to suppress multi-directional speech noise[J]. Journal of Tsinghua University(Science and Technology). 2020, 60(2): 181-188 https://doi.org/10.16511/j.cnki.qhdxxb.2019.21.022

参考文献

[1] ZENG F G, REBSCHER S, HARRISON W, et al. Cochlear implants:System design, integration, and evaluation[J]. IEEE Reviews in Biomedical Engineering, 2008, 1:115-142.
[2] BUECHNER A, DYBALLA K H, HEHRMANN P, et al. Advanced beamformers for cochlear implant users:Acute measurement of speech perception in challenging listening conditions[J]. PLoS One, 2014, 9(4):e95542.
[3] HU Y, LOIZOU P C. Environment-specific noise suppression for improved speech intelligibility by cochlear implant users[J]. The Journal of the Acoustical Society of America, 2010, 127(6):3689-3695.
[4] MENG Q L, ZHENG N H, Li X. Mandarin speech-in-noise and tone recognition using vocoder simulations of the temporal limits encoder for cochlear implants[J]. The Journal of the Acoustical Society of America, 2016, 139(1):301-310.
[5] GOEHRING T, BOLNER F, MONAGHAN J J M, et al. Speech enhancement based on neural networks improves speech intelligibility in noise for cochlear implant users[J]. Hearing Research, 2017, 344:183-194.
[6] CHEN Y S, GONG Q. Real-time spectrum estimation-based dual-channel speech-enhancement algorithm for cochlear implant[J]. Biomedical Engineering Online, 2012, 11:74.
[7] KONING R, MADHU N, WOUTERS J. Ideal time-frequency masking algorithms lead to different speech intelligibility and quality in normal-hearing and cochlear implant listeners[J]. IEEE Transactions on Biomedical Engineering, 2015, 62(1):331-341.
[8] KOKKINAKIS K, RUNGE C, TAHMINA Q, et al. Evaluation of a spectral subtraction strategy to suppress reverberant energy in cochlear implant devices[J]. The Journal of the Acoustical Society of America, 2015, 138(1):115-124.
[9] SPRIET A, VAN DEUN L, EFTAXIADIS K, et al. Speech understanding in background noise with the two-microphone adaptive beamformer BEAM^TM in the Nucleus Freedom^TM cochlear implant system[J]. Ear and Hearing, 2007, 28(1):62-72.
[10] 宫琴, 郑硕. 基于波束形成与最大似然估计的近距离双麦克风语音增强算法[J]. 清华大学学报(自然科学版), 2018, 58(6):603-608. GONG Q, ZHENG S. Beamforming and maximum likelihood estimationfor speech enhancement usingdual closely-spaced microphones[J]. Journal of Tsinghua University (Science and Technology), 2018, 58(6):603-608. (in Chinese)
[11] YOUSEFIAN N, LOIZOU P C. A dual-microphone speech enhancement algorithm based on the coherence function[J]. IEEE Transactions on Audio, Speech, and Language Processing, 2012, 20(2):599-609.
[12] AISSA-EL-BEY A, LINH-TRUNG N, ABED-MERAIM K, et al. Underdetermined blind separation of nondisjoint sources in the time-frequency domain[J]. IEEE Transactions on Signal Processing, 2007, 55(3):897-907.
[13] FU Q J, ZHU M, WANG X S. Development and validation of the Mandarin speech perception test[J]. The Journal of the Acoustical Society of America, 2011, 129(6):EL267-EL273.
[14] 陈又圣, 宫琴. 基于双TP型麦克风的电子耳蜗前端指向性语音增强系统的研制[J]. 仪器仪表学报, 2010, 31(9):1952-1958. CHEN Y S, GONG Q. Study on dual TP microphone-based directional speech enhancement system for cochlear implant[J]. Chinese Journal of Scientific Instrument, 2010, 31(9):1952-1958. (in Chinese)
[15] LOIZOU P C. Speech enhancement:Theory and practice[M]. Boca Raton, USA:CRC press, 2007.
[16] GOLDSWORTHY R L. Two-microphone spatial filtering improves speech reception for cochlear-implant users in reverberant conditions with multiple noise sources[J]. Trends in Hearing, 2014, 18:2331216514555489.