Speech Endpoint Detection Method Based on LMS Noise Reduction and Improved Dual-threshold

doi:10.16182/j.issn1004731x.joss.201709011

Abstract

Abstract: Endpoint detection of speech in low signal-to-noise ratio plays a very important role in voice processing. For a speech signal that is polluted by additive noise, it is possible to suppress the noise and keep the original speech relatively constant. A dual-threshold speech endpoint detection algorithm was proposed based on the least mean squares error (LMS) adaptive filtering. Dual median filtering smoothing was performed before and after double-parameter double-threshold detection. Through the Matlab simulation, the speech endpoint detection method was compared with other methods. In the noise environment with low signal-to-noise ratio, the endpoint detection effect of speech has better accuracy and robustness.

Key words: endpoint detection, LMS adaptive filtering, noise reduction, double threshold method, median filter, smoothing

CLC Number:

TN912.3

Zhu Chunli, Li Xin. Speech Endpoint Detection Method Based on LMS Noise Reduction and Improved Dual-threshold[J]. Journal of System Simulation, 2017, 29(9): 1950-1960.

References

[1] Dahl G E, Dong Yu, Li Deng, et al.Context-dependent Pre-trained Deep Neural Network for Large-vocabulary Speech Recognition[J]. IEEE Transactions on Audio, Speech, and Language Processing (S1558-7916), 2012, 20(1): 30-42.
[2] Rezayee A, Gazor S.An Adaptive KLT Approach for Speech Enhancement[J]. IEEE Transactions on Speech and Audio Processing (S1063-6676), 2001, 9(2): 87-95.
[3] Voran S, Catellier A.Multiple-description Speech Coding Using Speech-polarity Decomposition [C]// Proceedings of IEEE Global Telecommunications Conference. Washington DC, USA: IEEE Press, 2010: 1-6.
[4] Lee B K, Chang J H.Packet Loss Concealment Based on Deep Neural Networks for Digital Speech Transmission[J]. IEEE Transactions on Audio, Speech, and Language Processing (S2329-9290), 2016, 24(2): 378-387.
[5] 薛胜尧. 基于改进型双门限语音端点检测算法的研究[J]. 电子设计工程, 2015, 23(4): 78-81.
(Xue Shengrao.Research on speech endpoint detection based on the improved dual-threshold algorithm[J]. Electronic Design Engineering, 2015, 23(4): 78-81.)
[6] 邓艳容, 景新幸, 杨海燕, 等. 语音端点检测研究[J]. 计算机系统应用, 2012, 21(6): 240-243.
(Deng Yanrong, Jing Xinxing, Yang Haiyan, et al.Research on Endpoint Detection of Speech[J]. Computer Systems & Applications, 2012, 21(6): 240-243.)
[7] Boll S.Suppression of acoustic noise in speech using spectral subtraction[J]. IEEE Transactions on Acoustic, Speech and Signal Processing (S0096-3518), 1979, 27(2): 113-120.
[8] 孙一鸣, 吴杨扬, 李平. 基于改进双门限法的语音端点检测研究[J]. 长春理工大学学报(自然科学版), 2016, 39(1): 92-95.
(Sun Yiming, Wu Yangyang, Li Ping.Research on Speech Endpoint Detection Based on the Improved Dual-threshold[J]. Journal of Changchun University of Science and Technology (Natural Science Edition), 2016, 39(1): 92-95.)
[9] 马国栋, 阎树田, 贺成柱, 等. 基于LMS算法与RLS算法自适应滤波及仿真分析[J]. 电子设计工程, 2014, 22(6): 45-49.
(Ma Guodong, Yan Shutian, He Chengzhu, et al.Adaptive filtering and simulation analysis based on LMS and RLS algorithm[J]. Electronic Design Engineering, 2014, 22(6): 45-49.)
[10] 汪潮, 单家方. 一种新的变步长LMS自适应滤波算法仿真及性能分析[J]. 科技通报, 2015, 31(1): 87-89.
(Wang Chao, Shan Jiafang.Simulation of New Variable Step Size LMS Adaptive Filtering Algorithm[J]. Bulletin of Science and Technology, 2015, 31(1): 87-89.)
[11] 李善姬, 尹沧涛. 一种基于改进LMS算法的语音消噪方法[J]. 测控技术, 2010, 29(10): 35-37.
(Li Shanji, Yin Cangtao.A Voice De-Noise Method Based on the Adaptive LMS Algorithm[J]. Measurement Control Technology, 2010, 29(10): 35-37.)
[12] 郑珩, 宋斌. 基于小波阈值和LMS算法语音降噪算法研究[J]. 现代电子技术, 2015, 38(8): 37-40.
(Zheng Hang, Song Bin.Research on voice noise reduction algorithm based on LMS algorithm and wavelet threshold[J]. Modern Electronics Technique, 2015, 38(8): 37-40.)
[13] 徐文超, 王光艳, 耿艳香, 等. 基于谱减法和变步长LMS语音增强算法[J]. 计算机工程与应用, 2015, 51(1): 213-217.
(Xu Wenchao, Wang Guangyan, Geng Yanxiang, et al.Speech enhancement algorithm based on spectral subtraction and variable-step LMS algorithm[J]. Computer Engineering and Applications, 2015, 51(1): 213-217.)
[14] 李曼. 在Matlab中实现基于LMS算法语音信号去噪[J]. 电脑知识与技术, 2014, 10(32): 7649-7651.
(Li Man.Realization in Matlab Voice Signal Denoising Based on LMS Algorithm[J]. Computer Knowledge and Technology, 2014, 10(32): 7649-7651.)
[15] 纪振发, 杨晖, 李然, 等. 基于短时自相关及过零率的语音端点检测算法[J]. 电子科技, 2016, 29(9): 52-55.
(Ji Zhengfa, Yang Hui, Li Ran, et al.Speech Endpoint Detection Algorithm Based on Short Time Autocorrelation and Short-time Zero Crossing Rate[J]. Electronic Science and Technology, 2016, 29(9): 52-55.)
[16] 赵力. 语音信号处理 [M]. 北京: 机械工业出版社, 2003.
(Zhao Li.Voice Signal Processing [M]. Beijing, China: China Machine Press, 2003.)
[17] 王钟斐. 基于双门限-频带方差的语音端点检测方法研究[J]. 电子设计工程, 2016, 24(19): 86-88.
(Wang Zhongfei.Speech endpoint detection method research based on double threshold-frequency band variance[J]. Electronic Design Engineering, 2016, 24(19): 86-88.)