基于高斯混合模型的压缩域语音增强方法

doi:10.3969/j.issn.0372-2112.2012.10.022

电子学报 ›› 2012, Vol. 40 ›› Issue (10): 2031-2038.DOI: 10.3969/j.issn.0372-2112.2012.10.022

基于高斯混合模型的压缩域语音增强方法

梁岩, 鲍长春, 夏丙寅, 何玉文, 周璇, 李娜

北京工业大学电子信息与控制工程学院, 北京 100124

收稿日期:2012-05-21 修回日期:2012-08-06 出版日期:2012-10-25
- 作者简介:
- 梁岩女,1986年10月出生于山西朔州.2009年进入北京工业大学攻读硕士学位.主要研究方向:压缩域语音增强. E-mail:liangyan861003@163.com 鲍长春 男,1965年出生于内蒙古赤峰市.博士,教授,博士生导师.主要研究领域为语音与音频信号处理. E-mail:chchbao@bjut.edu.cn
- 基金资助:
- 北京市教育委员会科技发展计划重点项目 (No.KZ201110005005); 国家自然科学基金 (No.61072089); 北京市属高等学校人才强教计划

Compressed Domain Speech Enhancement Based on Gaussian Mixture Model

LIANG Yan, BAO Chang-chun, XIA Bing-yin, HE Yu-wen, ZHOU Xuan, LI Na

School of Electronic Information and Control Engineering, Beijing University of Technology, Beijing 100124, China

Received:2012-05-21 Revised:2012-08-06 Online:2012-10-25 Published:2012-10-25
- Supported by:
- Key Program of Science and Technology Development Plan of Beijing Municipality Education Commission (No.KZ201110005005); National Natural Science Foundation of China (No.61072089); Talents training program of higher education institutions in Beijing Municipality

摘要/Abstract

摘要： 为了有效利用纯净语音导抗谱频率参数(ISFs)的先验知识,本文针对ITU-T G.722.2宽带语音编码标准提出了一种基于高斯混合模型的压缩域语音增强方法.首先,将含噪语音、纯净语音的导抗谱频率参数,以及对应的增益调整因子构成特征矢量,并利用高斯混合模型拟合其概率密度;然后,在最小均方误差 (MMSE) 准则下对纯净语音的特征参数进行最优贝叶斯估计.为了兼容编码器中的非连续性传输模式,当处理信号为非语音信息时,算法在保持噪声帧谱包络参数不变的前提下,按固定比例调整对数帧能量;且若出现帧擦除情况,算法不调整接收到的码流,并按正常帧处理方式调整恢复后的参数以更新相关历史.本文采用ITU-T G.160标准进行了性能测试,结果表明,与参考方法相比,所提方法在保证信噪比提高能力的同时,可以达到更大的噪声衰减量,且增强语音的客观质量更优.

关键词: 语音增强, 参数域, 高斯混合模型, 贝叶斯估计, 非连续性传输, 帧擦除

Abstract: A Gaussian Mixture Model (GMM) based speech enhancement method in compressed domain used for ITU-T G.722.2 wideband speech codec is proposed to take full advantage of the prior knowledge of the Immittance Spectral Frequencies (ISFs) for the clean speech.Firstly,GMM is adopted to model the joint probability density of feature vectors which are composed by the ISFs of noisy speech and clean speech with the corresponding gain scaling factor.Secondly,an optimal Bayesian estimation of feature parameters derived from clean speech is obtained under the minimum mean square error (MMSE) criterion.To be compatible with the DTX (Discontinuous Transmission) mode,the logarithmic energy is attenuated and the ISFs remain when a SID (Silence Insertion Descriptor) frame is received.Furthermore,if ao erased frame is received,the bit stream is unchanged and the proposed method is performed on the recovered parameters for the memory update.The evaluation is conducted under the ITU-T G.160.The results indicate that,comparing with the reference method,the proposed method can produce larger amount of noise level reduction with better objective speech quality,while the SNR improvement remains acceptable.

Key words: speech enhancement, compressed domain, Gaussian Mixture Model, Bayesian estimation, DTX, frame erasure

中图分类号:

TN912.35

梁岩, 鲍长春, 夏丙寅, 等. 基于高斯混合模型的压缩域语音增强方法[J]. 电子学报, 2012, 40(10): 2031-2038.

LIANG Yan, BAO Chang-chun, XIA Bing-yin, et al. Compressed Domain Speech Enhancement Based on Gaussian Mixture Model[J]. Acta Electronica Sinica, 2012, 40(10): 2031-2038.

参考文献

[1] Ephraim Y.Speech enhancement using a minimum mean-square error log-spectral amplitude estimator [J].IEEE Transactions on Acoustics,Speech and Signal Processing,1985,33(2):443-445.
[2] Ephraim Y.Speech enhancement using a minimum mean-square error short-time spectral amplitude estimator [J].IEEE Transactions on Acoustics,Speech and Signal Processing,1984,32(6):1109-1121.
[3] R Chandran,D J Marchok.Compres domain noise reduction and echo suppression for network speech enhancement [A].Proceedings of the 43rd IEEE Midwest Symposium on Circuits and Systems [C].MI,USA:IEEE Press,2000.10-13.
[4] H Taddei,C Beaugeant,M de Meuleneire.Noise reduction on speech codec parameters [A].Proceedings of IEEE International Conference Acoustic Speech Signal Process (ICASSP) [C].NJ,USA:IEEE Press,2004.497-500.
[5] N Duetsch,H Taddei,C Beaugeant,T Fingscheidt.Noise reduction on speech codec parameters [A].Proceedings of IEEE International Conference on Acoustics,Speech,and Signal Processing (ICASSP '04) [C].USA:IEEE Press,2004,Vol.1.497-500.
[6] H M Goodarzi,S Seyedtabaii.Speech enhancement using spectral subtraction based on a modified noise minimum statistics estimation [A].Proceedings of the 2009 Fifth International Joint Conference on INC,IMS and IDC [C].Seoul,South Korea,2009.1339-1343.
[7] R A Sukkar,R C Younce,Peng Zhang.Method and Apparatus for Voice Quality Enhancement.USA:US2006215683-A1,2006-09-28.
[8] Cuntai Guan,Yongbin Chen,Boxiu Wu.Direct modulation on LPC coefficients with application to speech enhancement and improving the performance of speech recognition in noise [A].Proceedings of IEEE International Conference on Acoustics,Speech,and Signal Processing (ICASSP) [C].Minneapolis,MN,USA:IEEE Press,1993,Vol.2.107-110.
[9] E T Fapi,C Beauqeant,H Taddei,D Pastor.Noise reduction within network through modification of LPC parameters [A].Proceedings of the 7th International ITG Conference on Source and Channel Coding (SCC) [C].UIm,Germany,2008.1-6.
[10] P Harding,B Milner.Speech enhancement by reconstruction from cleaned acoustic features [A].Interspeech 2011 [C].Florence,Italy,2011.1189-1192.
[11] ITU-T Recommendation G.722.2.Wideband Coding of Speech at Around 16kbit/s Using Adaptive Multi-Rate Wideband (AMR-WB) [S].
[12] 刘鑫.宽带音频的非线性频带扩展技术研究 [D].北京:北京工业大学,2011.21-34. Liu Xin.Nonlinear Bandwidth Extending for Wideband Audio.Beijing:Beijing University of Technology,2011.21-34.(in Chinese)
[13] 胡广书.数字信号处理导论[M].北京:清华大学出版社,2005.294-296.
[14] ITU-T Recommendation G.722.2 Annex A.Comfort Noise Aspects [S].
[15] ITU-T Recommendation G.722.2 Annex B.Source Controlled Rate Operation [S].
[16] ITU-T Recommendation G.160.Voice Enhancement Devices for Mobile Networks [S].
[17] ITU-T Recommendation P.862.Perceptual Evaluation of Speech Quality (PESQ):An Objective Method for End-to-End Speech Quality Assessment of Narrow Band Telephone Networks and Speech Codes [S].
[18] ITU-T Recommendation G.722.2 Appendix I.Error Concealment of Erroneous or Lost Frames [S].

基于高斯混合模型的压缩域语音增强方法

Compressed Domain Speech Enhancement Based on Gaussian Mixture Model

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]	周静, 鲍长春, 张旭. 基于聚焦信号子空间估计导向矢量的干扰声源抑制方法[J]. 电子学报, 2023, 51(1): 76-85.
[2]	李凯, 张可心. 结构α-熵的加权高斯混合模型的子空间聚类[J]. 电子学报, 2022, 50(3): 718-725.
[3]	欧世峰, 赵艳磊, 宋鹏, 高颖. 基于概率耦合的双直接判决先验信噪比估计算法[J]. 电子学报, 2020, 48(8): 1605-1614.
[4]	袁文浩, 胡少东, 时云龙, 李钊, 梁春燕. 一种用于语音增强的卷积门控循环网络[J]. 电子学报, 2020, 48(7): 1276-1283.
[5]	石雪, 李玉, 赵泉华. 自适应类别的层次高斯混合模型遥感影像分割[J]. 电子学报, 2020, 48(1): 131-136.
[6]	袁文浩, 娄迎曦, 梁春燕, 夏斌. 利用生成噪声提高语音增强方法的泛化能力[J]. 电子学报, 2019, 47(4): 791-797.
[7]	陈楠, 鲍长春. 基于双耳线索编码原理的语音增强方法[J]. 电子学报, 2019, 47(1): 227-233.
[8]	赖裕平, 高宁, 何闻达, 平原, 杜春来, 王宝成, 丁洪伟. 贝塔混合模型的变分贝叶斯学习及应用[J]. 电子学报, 2018, 46(7): 1787-1792.
[9]	练秋生, 侯亚伟, 苏月明, 石保顺. 基于高斯混合模型的衍射成像算法[J]. 电子学报, 2018, 46(10): 2347-2350.
[10]	袁文浩, 梁春燕, 夏斌, 孙文珠. 一种融合相位估计的深度卷积神经网络语音增强方法[J]. 电子学报, 2018, 46(10): 2359-2366.
[11]	白海钏, 鲍长春, 刘鑫. 基于局部最小二乘支持向量机的音频频带扩展方法[J]. 电子学报, 2016, 44(9): 2203-2210.
[12]	车滢霞, 俞一彪. 约束条件下的结构化高斯混合模型及非平行语料语音转换[J]. 电子学报, 2016, 44(9): 2282-2288.
[13]	栗雨晴, 礼欣, 韩煦, 宋丹丹, 廖乐健. 基于双语词典的微博多类情感分析方法[J]. 电子学报, 2016, 44(9): 2068-2073.
[14]	张宏欣, 周穗华, 冯士民. 高斯粒子流滤波器[J]. 电子学报, 2016, 44(4): 795-803.
[15]	孟宪波, 鲍长春. 基于最小控制GARCH模型的噪声估计算法[J]. 电子学报, 2016, 44(3): 747-752.