Speech Endpoint Detection Algorithm Based on MFCC Distance in Complex Noise

doi:10.19678/j.issn.1000-3428.0053966

Computer Engineering ›› 2020, Vol. 46 ›› Issue (3): 309-314. doi: 10.19678/j.issn.1000-3428.0053966

• Development Research and Engineering Application • Previous Articles Next Articles

Speech Endpoint Detection Algorithm Based on MFCC Distance in Complex Noise

HAN Yunxiao¹, SHAO Qing¹, FU Yuxiang², GUO Qing²

1. School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China;
2. The 36 th Research Institute of China Electronics Technology Group Corporation, Jiaxing, Zhejiang 314000, China

Received:2019-02-21 Revised:2019-04-20 Published:2019-03-27

复杂噪声中基于MFCC距离的语音端点检测算法

韩云霄¹, 邵清¹, 符玉襄², 郭庆²

1. 上海理工大学光电信息与计算机工程学院, 上海 200093;
2. 中国电子科技集团公司第三十六研究所, 浙江嘉兴 314000

作者简介:韩云霄(1993-),女,硕士,主研方向为自然语言处理;邵清,副教授、博士;符玉襄,高级工程师、博士;郭庆,工程师、硕士。
基金资助:
国家自然科学基金（61703278）；上海市科委科研计划（17511107203）。

Abstract

Abstract: To improve the accuracy of speech signal endpoint detection under complex noise environment,this paper proposes a multidimensional feature speech signal endpoint detection algorithm based on MFCC distance.By calculating the MFCC distance of the speech signal and combining short time energy and short time over zero rate,this algorithm corrects the feature distance,updates the threshold value and establishes the adaptive noise model to achieve the speech signal endpoint detection in complex noise.Experimental results show that under the condition of same calculation efficiency,the proposed algorithm has higher detection accuracy compared with the two classic detection algorithms based on double threshold energy and cepstrum distance.

Key words: speech signal, endpoint detection, multidimensional features, MFCC distance, adaptive noise model

摘要： 为提高复杂噪声环境下语音信号端点检测的准确率，提出一种基于梅尔频谱倒谱系数（MFCC）距离的多维特征语音信号端点检测算法。通过计算语音信号的MFCC距离，结合短时能量和短时过零率对特征距离进行修正，并更新其阈值，建立自适应噪声模型，实现复杂噪声中语音信号端点的准确检测。实验结果表明，与基于双门限能量和基于倒谱距离的2种经典检测算法相比，在计算效率相同的条件下，该算法的检测准确率更高。

关键词: 语音信号, 端点检测, 多维特征, 梅尔频谱倒谱系数距离, 自适应噪声模型

CLC Number:

TP37

HAN Yunxiao, SHAO Qing, FU Yuxiang, GUO Qing. Speech Endpoint Detection Algorithm Based on MFCC Distance in Complex Noise[J]. Computer Engineering, 2020, 46(3): 309-314.

韩云霄, 邵清, 符玉襄, 郭庆. 复杂噪声中基于MFCC距离的语音端点检测算法[J]. 计算机工程, 2020, 46(3): 309-314.

/ / Recommend / Download Citations

URL: http://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0053966

http://www.ecice06.com/EN/Y2020/V46/I3/309

Figures/Tables 5

References

[1] AGARWALLA S,SARMA K K.Machine learning based sample extraction for automatic speech recognition using dialectal Assamese speech[J].Neural Networks,2016,78:97-111.
[2] DESAI N P,LEHMAN C,MUNSON B,et al.Supervised and unsupervised machine learning approaches to classifying chimpanzee vocalizations[J].The Journal of the Acoustical Society of America,2018,143(3):1786-1786.
[3] MONAGHAN J J M,GOEHRING T,YANG X,et al.Auditory inspired machine learning techniques can improve speech intelligibility and quality for hearing-impaired listeners[J].The Journal of the Acoustical Society of America,2017,141(3):1985-1998.
[4] FAYEK H M,LECH M,CAVEDON L.Evaluating deep learning architectures for speech emotion recognition[J].Neural Networks,2017,92:60-68.
[5] WU Bo,LI Kehuang,GE Fengpei,et al.An end-to-end deep learning approach to simultaneous speech dereverberation and acoustic modeling for robust speech recognition[J].IEEE Journal of Selected Topics in Signal Processing,2017,11(8):1289-1300.
[6] LEE J,SKOGLUND J,SHABESTARY T,et al.Phase-sensitive joint learning algorithms for deep learning-based speech enhancement[J].IEEE Signal Processing Letters,2018,25(8):1276-1280.
[7] ZHOU Yan,LIU Tao,SHANG Li.Speech sparse decomposition algorithm based on immune matching pursuit[J].Computer Engineering,2012,38(21):161-163,167.(in Chinese)周燕,刘韬,尚丽.基于免疫匹配追踪的语音稀疏分解算法[J].计算机工程,2012,38(21):161-163,167.
[8] ABDELKHALIK O,DARANI S.Evolving hidden genes in genetic algorithms for systems architecture optimization[J].Journal of Dynamic Systems,Measurement,and Control,2018,140(10):101015-101026.
[9] TANG Min,JIN Jian,LIU Ying,et al.Integrating topic,sentiment,and syntax for modeling online reviews:a topic model approach[J].Journal of Computing and Information Science in Engineering,2019,19(1):6-20.
[10] WANG Bo,YU Fengqin.Speech endpoint detection based on multi-scale sample entropy and threshold[J].Computer Engineering,2016,42(12):268-271.(in Chinese)王波,于凤芹.基于多尺度样本熵与阈值的语音端点检测[J].计算机工程,2016,42(12):268-271.
[11] XIAO Xi,ZHOU Lu.Speech recognition adaptive clustering feature extraction algorithms based on the k-means algorithm and the normalized intra-class variance[J].Journal of Tsinghua University(Science and Technology),2017,57(8):857-861.(in Chinese)肖熙,周路.基于k均值和基于归一化类内方差的语音识别自适应聚类特征提取算法[J].清华大学学报(自然科学版),2017,57(8):857-861.
[12] 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):1950-1960,1967.(in Chinese)朱春利,李昕.基于LMS减噪与改进的双门限语音端点检测方法[J].系统仿真学报,2017,29(9):1950-1960,1967.
[13] SHIN W H,LEE B S,LEE Y K,et al.Speech/non-speech classification using multiple features for robust endpoint detection[C]//Proceedings of 2000 IEEE International Conference on Acoustics,Speech,and Signal Processing.Washington D.C.,USA:IEEE Press,2000:1399-1402.
[14] ZHANG Long,XU Xu,CHEN Huang,et al.Supervised single-channel speech dereverberation and denoising using a two-stage model based sparse representation[J].Speech Communication,2018,97:1-8.
[15] VIEN Q T,NGUYEN H X,TRESTIAN R,et al.A hybrid double-threshold based cooperative spectrum sensing over fading channels[J].IEEE Transactions on Wireless Communications,2015,15(3):1821-1834.
[16] DUSTOR A.Influence of noise and voice activity detection on speaker verification[C]//Proceedings of International Conference on Computer Networks.Berlin,Germany:Springer,2016:207-215.
[17] SCHAFER R W.Homomorphic systems and cepstrum analysis of speech[M].Berlin,Germany:Springer,2008.
[18] ZHANG Rui,HU Ruimin,LI Gang,et al.Spectral tilt estimation for speech intelligibility enhancement using RNN based on all-pole model[C]//Proceedings of International Conference on Multimedia Modeling.Berlin,Germany:Springer,2019:144-156.
[19] REN Yanzhen,LIU Dengkai,YANG Jing,et al.An AMR adaptive steganographic scheme based on the pitch delay of unvoiced speech[J].Multimedia Tools and Applications,2019,78:8091-8111.

Please choose a citation manager

Content to export