一种移动音频编码自适应丢帧隐藏算法  

An Adaptive Frame Loss Concealment Algorithm for Mobile Audio Coding

作　　者：项慨[1,2] 胡瑞敏[1] 

机构地区：[1]武汉大学计算机学院国家多媒体软件工程技术研究中心,湖北武汉430072 [2]湖北经济学院信息管理学院,湖北武汉430205

出　　处：《武汉大学学报（理学版）》2014年第5期466-470,共5页Journal of Wuhan University:Natural Science Edition

基　　金：国家重大科技专项资助项目(2010ZX03004-003-03);国家自然科学基金重点项目(61231015);国家自然科学基金面上青年项目(61102127;61201340;61201169;61272278);湖北省教育厅青年项目(Q20132207)

摘　　要：针对主流移动音频编码器AVS-P10所采用的语音丢帧隐藏算法在不同丢帧的情况下语音恢复效果存在的不足,提出了一种自适应的语音丢帧隐藏算法,该算法根据不同的坏帧模式自适应地采取不同的坏帧恢复方法,对坏帧中的基音周期参数和ISF系数充分利用前后分别收到的好帧采用更优化的线性预测方法恢复.实验表明:与AVS-P10中的丢帧隐藏算法相比,本文方法在坏帧率低于3%时有更好的语音质量恢复效果.For defects of the traditional frame loss concealment algorithm employed in the current mainstream audio codec AVS-P10 standard under different circumstances of dropped frames, a new adaptive frame loss concealment algorithm was presented to solve the problem. The algorithm adaptively took different recovery method based on a va- riety of different bad frame modes, and the pitch period parameters and 1SF coefficients in had frame are recovered by a more optimized linear prediction method taking advantage of relative parameters in good frames received respectively before or after the current bad frame. The experiment results proved that the proposed method had better recovery effect than original method employed in audio codec of AVS-P10 standard when the frame error rate was less than 3 percentages of all transmitted frames.

关 键 词：移动音频编码 丢帧隐藏 差错掩蔽 音频信号处理 

分 类 号：TN912.3[电子电信—通信与信息系统]