利用两级时域联合的包层语音质量评价模型

doi:10.3969/j.issn.1001-2400.2013.03.003

J4 ›› 2013, Vol. 40 ›› Issue (3): 14-19+94.doi: 10.3969/j.issn.1001-2400.2013.03.003

利用两级时域联合的包层语音质量评价模型

江亮亮;杨付正;任光亮

(西安电子科技大学综合业务网理论及关键技术国家重点实验室，陕西西安 710071)

收稿日期:2012-10-08 出版日期:2013-06-20 发布日期:2013-07-29
作者简介:江亮亮(1988-)，男，西安电子科技大学博士研究生，E-mail: lljiang@stu.xidian.edu.cn．
基金资助:
国家自然科学基金资助项目(60902081,60902052);高等学校学科创新引智计划资助项目(B08038)

Packet-layer model for voice quality assessment using two-level temporal pooling scheme

JIANG Liangliang;YANG Fuzheng;REN Guangliang

(State Key Lab. of Integrated Service Networks, Xidian Univ., Xi'an 710071， China)

Received:2012-10-08 Online:2013-06-20 Published:2013-07-29

摘要/Abstract

摘要：

针对相同丢包率下不同丢包模式对应的语音质量存在差异的情况，提出了一种能够反映丢包模式对语音质量影响的包层语音质量评价模型．首先通过分析数据包头获取编码速率和丢包位置等信息，在此基础上，结合静音检测技术及误码传播结果预测每一帧的质量；然后根据人的感知特性将语音序列自由划分为变长帧组，并联合各帧的质量得到帧组质量；最后，综合各帧组的质量得到语音序列的总质量．提出的模型在两级时域联合过程中，为失真严重的区域分配更大的权重，从而有效反映丢包模式对语音质量的影响．实验结果表明，相比于国际标准G.107中的E-model，所提模型的评分与语音质量感知评估方法的评分相比，皮尔森相关系数平均提高了0.0129，同时均方根误差平均降低了0.0234．

关键词: 语音质量评价, 时域联合, 丢包, 服务质量

Abstract:

Aiming at the problem that the voice qualities corresponding to different packet loss patterns show significant differences at the same packet loss rate, a packet-layer model for voice quality assessment, which well reflects the effect of the packet loss patterns on the voice quality, is presented. First, the information about the coding bit-rate and packet loss is obtained by analyzing the packet header, on the basis of which the frame quality is measured with the further information about silence detection and error propagation. Then the voice sequence is divided into groups of frames (GOFs) with a variable length and a short-term temporal pooling method is employed to obtain the GOF quality. Finally, the overall voice quality is determined by the long-term temporal pooling of the GOF qualities. The proposed two-level temporal pooling scheme well describes the effect of different packet loss patterns on the voice quality since the strongest impairments are predominately emphasized. Experimental results show that the presented model can lead to an increment of about 0.0129 in the Pearson Correlation coefficient (PCC) and a decrement of about 0.0234 in the Root Mean Squared Error (RMSE) compared with the E-model in ITU-T recommendation G.107.

Key words: voice quality assessment, temporal pooling, packet loss, quality of service

中图分类号:

TN912

江亮亮;杨付正;任光亮. 利用两级时域联合的包层语音质量评价模型[J]. J4, 2013, 40(3): 14-19+94.

JIANG Liangliang;YANG Fuzheng;REN Guangliang. Packet-layer model for voice quality assessment using two-level temporal pooling scheme[J]. J4, 2013, 40(3): 14-19+94.

参考文献

［1］ Daengsi T, Preechayasomboon A, Wutiwiwatchai C, et al. A Study of VoIP Quality Evaluation: User Perception of Voice Quality from G.729, G. 711 and G. 722 ［C］//Proc of the 9th Annual IEEE Consumer Communications and Networking Conference. Piscataway: IEEE Computer Society, 2012: 342-345.
［2］ Jelassi S, Rubino G, Melvin H, et al. Quality of Experience of VoIP Service: a Survey of Assessment Approaches and Open Issues ［J］. IEEE Communications Surveys ＆ Tutorials, 2012, 14(2): 491-513.
［3］ Rix A W, Beerends J G, Kim D S, et al. Objective Assessment of Speech and Audio Quality: Technology and Applications ［J］. IEEE Trans on Audio, Speech, and Language Processing, 2006, 14(6): 1890-1901.
［4］ ITU-T Recommendation P.861. Objective Quality Measurement of Telephone-band (300-3400 Hz) Speech Codecs ［S］. Geneva: Telecommunication Standardization Sector, 1996.
［5］ ITU-T Recommendation P.862. Perceptual Evaluation of Speech Quality (PESQ), an Objective Method for End to End Speech Quality Assessment of Narrowband Telephone Networks and Speech Codecs ［S］. Geneva: Telecommunication Standardization Sector, 2001.
［6］ Moller S, Chan W, Cote N, et al. Speech Quality Estimation: Models and Trends ［J］. IEEE Signal Processing Magazine, 2011, 28(6): 18-28.
［7］ Narwaria M, Lin W, McLoughlin I V, et al. Nonintrusive Quality Assessment of Noise Suppressed Speech with Mel-Filtered Energies and Support Vector Regression ［J］. IEEE Trans on Audio, Speech, and Language Processing, 2012, 20(4): 1217-1232.
［8］ Takahashi A, Yoshino H, Kitawaki N. Perceptual QoS Assessment Technologies for VoIP ［J］. IEEE Communications Magazine, 2004, 42(7): 28-34.
［9］ Clark A. Modeling the Effects of Burst Packet Loss and Recency on Subjective Voice Quality ［C］//Proc of IP-Telephony Workshop. New York: IEEE, 2001: 123-127.
［10］ ITU-T SG12 Temporary Document TD 297. Updated Draft Terms of Reference for P.NAMS ［S］. Geneva: Telecommunication Standardization Sector, 2010.
［11］ ITU-T Recommendation G.107. The E-model, a Computational Model for Use in Transmission Planning ［S］. Geneva: Telecommunication Standardization Sector, 2002.
［12］ Jelassi S, Rubino G. A Comparison Study of Automatic Speech Quality Assessors Sensitive to Packet Loss Burstiness ［C］//Proc of the 8th Annual IEEE Consumer Communications and Networking Conference. Piscataway: IEEE Computer Society, 2011: 415-420.
［13］ Yang Fuzheng, Jiang Liangliang, Li Xiao. Real-time Quality Assessment for Voice over IP ［J］. Concurrency and Computation: Practice and Experience, 2012, 24(11): 1192-1199.
［14］ Radhakrishnan K, Larijani H, Buggy T. A Non-intrusive Method to Assess Voice Quality over Internet ［C］//Proc of the 2010 International Symposium on Performance Evaluation of Computer and Telecommunication Systems. Piscataway: IEEE Computer Society, 2010: 380-386.
［15］李维, 杨付正. 考虑包内容特性的网络语音质量评价模型［J］. 西安电子科技大学学报, 2011, 38(2): 23-28.
Li Wei, Yang Fuzheng. No-reference Quality Assessment Model for Networked Speech Based on the Content Feature of Packets ［J］. Journal of Xidian University, 2011, 38(2): 23-28.
［16］ Rohaly A M, Lu J, Franzen N R, et al. Comparison of Temporal Pooling Methods for Estimating the Quality of Complex Video Sequences ［C］//Proc of SPIE Human Vision and Electronic Imaging Conference. Bellingham: SPIE, 1999: 218-226.
［17］ Rix A, Beerends J, Hollier M, et al. Perceptual Evaluation of Speech Quality (PESQ)-a New Method for Speech Quality Assessment of Telephone Networks and Codecs ［C］//Proc of IEEE International Conference on Acoustics, Speech, and Signal Processing. Salt Lake City: IEEE, 2001: 749-752.
［18］ ITU Rep. COM12-D97-E. Packet Loss Distributions and Packet Loss Models ［S］. Geneva: ITU-T Study Group 12, 2003.

[1]	韩家曦, 赵辉, 冯南之, 王静, 万波, 王泉. 边缘计算平台中面向分簇的半在线调度方法[J]. 西安电子科技大学学报, 2024, 51(6): 104-116.
[2]	曹静;武君胜;杨文超;王硕晨. 认知无线ad hoc网的多尺度跨层路由协议[J]. 西安电子科技大学学报, 2018, 45(2): 77-83.
[3]	段瑞猛;马文平;罗维;赵飞飞. 一个毫微微小区网络中的全分布式资源分配策略[J]. 西安电子科技大学学报, 2017, 44(3): 36-42.
[4]	宋佳润;徐子强;杨付正. 结合网络视频传输特性的参数规划模型[J]. 西安电子科技大学学报, 2016, 43(2): 168-173.
[5]	杨鹏;李祥攀;刘豆 . LTE中增强视频业务QoS的时延感知调度算法 [J]. J4, 2015, 42(2): 140-145.
[6]	孟宪佳;马建峰;卢笛;王一川. 面向可信和行为模式匹配的两层服务选择方法[J]. J4, 2014, 41(4): 198-204.
[7]	徐琳;曲仕茹;鱼滨. ITSGrid资源调度策略的研究[J]. J4, 2014, 41(3): 203-208.
[8]	彭飞;邓浩江;刘磊. 面向个性化服务推荐的QoS动态预测模型[J]. J4, 2013, 40(4): 174-180.
[9]	刘河潮;常义林;元辉;杨家玮. 一种网络丢包的无参考视频质量评估方法[J]. J4, 2012, 39(2): 29-34.
[10]	杨延飞;朱樟明;周端;杨银堂. 用于片上网络的延时无关异步动态优先级仲裁器[J]. J4, 2012, 39(1): 42-48+110.
[11]	李维;杨付正. 考虑包内容特性的网络语音质量评价模型[J]. J4, 2011, 38(2): 23-28+98.
[12]	张琰;盛敏;李建东;田野;姚俊良. 一种新型动态协同节点选择策略[J]. J4, 2010, 37(1): 73-79.
[13]	黄伯虎;段振华. 量子遗传算法在Web服务选择中的应用[J]. J4, 2010, 37(1): 56-61+67.
[14]	李建东;陈婷;邓少平;李长乐. WiMAX中保证QoS的跨层架构研究[J]. J4, 2009, 36(4): 573-582.
[15]	张国鹏;张海林. 无线局域网中业务流接入控制的博弈论算法 [J]. J4, 2008, 35(5): 805-810.

利用两级时域联合的包层语音质量评价模型

Packet-layer model for voice quality assessment using two-level temporal pooling scheme

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

Metrics

本文评价

推荐阅读 0