基于长时信息的自适应话音激活检测

doi:10.3969/j.issn.0372-2112.2018.04.016

电子学报 ›› 2018, Vol. 46 ›› Issue (4): 878-885.DOI: 10.3969/j.issn.0372-2112.2018.04.016

基于长时信息的自适应话音激活检测

杨绪魁, 屈丹, 张文林, 闫红刚

解放军信息工程大学, 河南郑州 450001

收稿日期:2016-11-16 修回日期:2017-04-24 出版日期:2018-04-25
- 作者简介:
- 杨绪魁 男,1988年10月出生,福建光泽人.现为解放军信息工程大学在读博士研究生,研究方向为语音信号处理与识别、机器学习等.E-mail:gzyangxk@163.com;屈丹女,1974年9月出生,吉林九台人,现为解放军信息工程大学信息系统工程副教授、博士生导师.主要研究方向为语音信号处理与识别、人工智能等;张文林 男,1982年11月出生,河北蕲春人.现为解放军信息工程大学信息系统工程学院讲师.主要研究方向语音信号处理与识别、人工智能等;闫红刚 男,1975年10月出生,河南驻马店人.现为解放军信息工程大学信息系统工程学院副教授.主要研究方向通信信号分析、语音处理与识别、机器学习等
- 基金资助:
- 国家自然科学基金 (No.61673395，No.61403415）; 河南省自然科学基金 (No.162300410331）

Adaptive Voice Activity Detection Based on Long-Term Information

YANG Xu-kui, QU Dan, ZHANG Wen-lin, YAN Hong-gang

PLA Information Engineering University, Zhengzhou, Henan 450001, China

Received:2016-11-16 Revised:2017-04-24 Online:2018-04-25 Published:2018-04-25

摘要/Abstract

摘要： 语音信号的长时信息应用于话音激活检测中表现优越.利用三种听觉滤波器组，对语音信号进行非线性的谱分解，本文提出了六种基于听觉滤波器组的长时信息，并提出了基于长时信息的自适应话音激活检测算法.该算法无需训练数据，根据多种长时信息，直接在待测信号中挑选出类别明确的信号，然后利用这些信号训练分类模型，对待测信号按帧进行语音-非语音分类.在TIMIT语音库和NOISEX-92噪声库上的实验表明，该算法在极低信噪比环境下，仍表现出更高的准确性和更强的稳健性.同时，在线实验表明，算法在实时处理中仍能取得优异的性能.

关键词: 话音激活检测, 长时信息, 听觉滤波器, 自适应

Abstract: The long-term information of speech signals shows excellent performances in the applications of voice activity detection.Six types of long-term information based on auditory filter banks are proposed through the non-linear spectral decomposition with three different auditory filters.Further,an adaptive voice activity detection algorithm based on these types of long-term information is proposed.Without additional training data,this algorithm use the data selecting from the test signals according to long-term information to train a speech/non-speech classifier,and classifies the current test signals using the speech/non-speech classifier frame by frame.Experiments on TIMIT dataset and NOISEX-92 dataset show that the algorithm improves the performance of VAD with higher accuracy and stronger robustness in low SNR noisy environments.The online experiments show that it can also obtain a good performance in real-time processing conditions.

Key words: voice activity detection, long-term information, auditory filter bank, adaptive

中图分类号:

TN912.34

杨绪魁, 屈丹, 张文林, 闫红刚. 基于长时信息的自适应话音激活检测[J]. 电子学报, 2018, 46(4): 878-885.

YANG Xu-kui, QU Dan, ZHANG Wen-lin, YAN Hong-gang. Adaptive Voice Activity Detection Based on Long-Term Information[J]. Acta Electronica Sinica, 2018, 46(4): 878-885.

参考文献

[1] RAMIREZ J,GORRIZ J-M,SEGURA J-C.Voice activity detection,fundamentals and speech recognition system robustness,robust speech recognition and understanding[OL].https://www.intechopen.com/books/robust_speech_ecognition_and_understanding/voice_activety_detection_fundamentals_and_speech_recognition_system_robustness, 2016-11-16.
[2] WISDOM S,OKOPAL G,ATLAS L,PITTON J.Voice activity detection using subband noncircularity[A].Proceedings of IEEE International Conference on Acoustics,Speech,and Signal Processing (ICASSP)[C].Brisbane,Australia,2015.4505-4509.
[3] HEESE F,NIERMANN M,VARY P.Speech-codebook based soft voice activity detection[A].Proceedings of IEEE International Conference on Acoustics,Speech,and Signal Processing (ICASSP)[C].Brisbane,Australia,2015.4335-4339.
[4] TAO F-J,HANSEN H-L,BUSSO C.An unsupervised visual-only voice activity detection approach using temporal orofacial features[A].Proceedings of 16th Annual Conference of the International Speech Communication Association (INTERSPEECH)[C].Dresden,Germany,2015.2302-2306.
[5] ZHAN G,HUANG Z-Q,et al.Spectrographic speech mask estimation using the time-frequency correlation of speech presence[A].Proceedings of 16th Annual Conference of the International Speech Communication Association (INTERSPEECH)[C].Dresden,Germany,2015.2287-2291.
[6] RAMIREZ J,SEGURA J-C,BENITEZ C,et al.Efficient voice activity detection algorithms using long-term speech information[J].Speech Communication,2004,42(3):271-287.
[7] GHOSH P-K,TSIARTAS A,NARAYANAN S.Robust voice activity detection using long-term signal variability[J].IEEE Transactions on Audio,Speech,and Language Processing,2011,19(3):600-613.
[8] MA Y,NISHIHARA A.Efficient voice activity detection algorithm using long-term spectral flatness measure[J]. EURASIP Journal on Audio,Speech and Music Processing,2013:87,DOI:10.1186/1687-4722-2013-21.
[9] YANG X-K,He L,QU D,ZHANG W-Q.Voice activity detection algorithm based on long-term pitch information[J].EURASIP Journal on Audio,Speech,and Music Processing,2016:14,DOI:10.1186/s13636-016-0092-y.
[10] DAVIS S,MERMELSTEIN P.Comparison of parametric representations for monosyllabic word recognitions in continuously spoken sentences[J].IEEE Transactions on Acoustics,Speech and Signal Processing,1980,28(4):357-366.
[11] SCHLUTER R,BEZRUKOV I,WAGNER H,NEY H.Gammatone features and feature combination for large vocabulary speech recognition[A].Proceedings of IEEE International Conference on Acoustics,Speech,and Signal Processing (ICASSP)[C].Hawaii,USA:IEEE,2007.649-652.
[12] MEINARD Muller.Information Retrieval for Music and Motion[M].Berlin:Springer Verlag,2007.51-55.
[13] SEGBROECH M-V,TSIARTAS A,NARAYANAN S-S.A robust frontend for VAD:exploiting contextual,discriminative and spectral cues of human voice[A].Proceedings of 14th Annual Conference of the International Speech Communication Association (INTERSPEECH)[C].Lyon,France,2013.704-708.
[14] KINNUNEN T,RAJAN P.A practical,self-adaptive voice activity detector for speaker verification with noise telephone and microphone data[A].Proceedings of IEEE International Conference on Acoustics,Speech,and Signal Processing (ICASSP)[C].Vancouver,Canada,2013.7229-7233.
[15] Georgiou T-T.Distances Between Power Spectral Densities[R].Technique Report,arXiv:math/0607026v2,2006.
[16] JOHANNESMA P-I-M.The pre-response stimulus ensemble of neurons in the cochlear nucleus[A].Proceedings of IPO Symposium on Hearing Theory[C].Eindhoven,Netherlands,1972.58-69.
[17] GERKMANN T,RICHARD C-H.Unbiased MMSE-based noise power estimation with low complexity and low tracking delay[J].IEEE Transactions on Audio,Speech,and Language Processing,2012,20(4):1383-1393.
[18] GAROFOLO J-S,LAMEL L-F,Fisher W-M,et al.TIMIT Acoustic-Phonetic Continuous Speech Corpus[R].NIST Interagency/Internal Report (NISTIR)-4930,1993.
[19] VARGA A,STEENEKEN H-J-M.Assessment for automatic speech recognition:Ii.NOISEX-92:a database and an experiment to study the effect of additive noise on speech recognition systems[J].Speech Communication,1993,12(3):247-251.
[20] SOHN J,KIM N-S,SUNG W.A statistical model-based voice activity detection[J].IEEE Signal Processing Letters,1999,6(1):1-3.
[21] TAN L-N,BORGSTROM B-J,ALWAN A.Voice activity detection using harmonic frequency components in likelihood ratio Test[A].IEEE International Conference on Acoustics,Speech,and Signal Processing (ICASSP)[C].Dallas,USA:IEEE,2010.4466-4469.

基于长时信息的自适应话音激活检测

Adaptive Voice Activity Detection Based on Long-Term Information

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	陇盛, 陶蔚, 张泽东, 陶卿. 非光滑强凸情形Adam型算法的最优收敛速率[J]. 电子学报, 2022, 50(9): 2049-2059.
[2]	孔玮, 刘云, 李辉, 崔雪红, 杨浩冉. 基于全局自适应有向图的行人轨迹预测[J]. 电子学报, 2022, 50(8): 1905-1916.
[3]	钟沛龙, 黎明, 何超, 陈昊. 基于SOM聚类和自适应算子选择的高维多目标进化算法[J]. 电子学报, 2022, 50(8): 1959-1974.
[4]	江兵兵, 何文达, 吴兴宇, 项俊浩, 洪立斌, 盛伟国. 基于自适应图学习的半监督特征选择[J]. 电子学报, 2022, 50(7): 1643-1652.
[5]	易令, 李泽平. 基于深度强化学习的码率自适应算法研究[J]. 电子学报, 2022, 50(5): 1192-1200.
[6]	李纪三, 纪彦星, 曹鼎, 刘溶, 任渊. 基于广义时间窗的旋转相控阵雷达资源调度算法[J]. 电子学报, 2022, 50(5): 1050-1057.
[7]	姜奇, 文悦, 张瑞杰, 魏福山, 马建峰. 面向智能手机的自适应触屏持续认证方案[J]. 电子学报, 2022, 50(5): 1131-1139.
[8]	杨曦, 张鑫, 郭浩远, 王楠楠, 高新波. 基于不变特征的多源遥感图像舰船目标检测算法[J]. 电子学报, 2022, 50(4): 887-899.
[9]	王延达, 陈炜通, 皮德常, 岳琳. 一种自适应记忆神经网络多跳读取与覆盖度机制结合的药物推荐模型[J]. 电子学报, 2022, 50(4): 943-953.
[10]	吴孙勇, 李东升, 薛秋条, 孙希延, 蔡如华. 基于PHD滤波的雷达起伏目标检测前跟踪算法研究[J]. 电子学报, 2022, 50(3): 691-702.
[11]	李志博, 李清宝, 兰明敬, 孙剑帆. 基于镜像选择序优化的MART算法[J]. 电子学报, 2022, 50(2): 314-325.
[12]	董明, 李敬, 索永录, 唐恩贤, 马宏伟, 陈渊, 张广明, 万翔. 基于改进果蝇优化算法正交匹配追踪的超声信号降噪方法[J]. 电子学报, 2022, 50(2): 493-501.
[13]	李双明, 关欣, 王海滨. 参数自适应的析取云模糊置信规则识别方法[J]. 电子学报, 2022, 50(2): 396-403.
[14]	王洪雁, 周贺. 改善MIMO-STAP检测性能的收发空时资源配置方法[J]. 电子学报, 2022, 50(11): 2619-2628.
[15]	高雷阜, 张梦瑶, 赵世杰. 融合簇边界移动与自适应合成的混合采样算法[J]. 电子学报, 2022, 50(10): 2517-2529.