RASTA滤波在语音通信质量客观评价中应用的研究  

A Study for the Application of RASTA on Objective Communication Speech Quality Evaluation

作　　者：王炜[1] 刘峰[1] 吴淑珍[1] 

机构地区：[1]北京大学信息科学技术学院,北京100871

出　　处：《北京大学学报（自然科学版）》2003年第5期697-702,共6页Acta Scientiarum Naturalium Universitatis Pekinensis

摘　　要：介绍了在语音通信质量客观评价中使用RASTA(RelAtiveSpecTrA)滤波的研究结果。通过分别将RASTA滤波与感知线性预测 (PLP ,Perceptually Linear Prediction)方法和美倒谱 (MFCC ,Mel Frequency Cepstral Coefficient)方法相结合对通过短波或超短波信道的男女生文件进行拟合研究 ,均取得了比较好的效果。与PLP方法结合关系数提高了 12 % ,与MFCC方法结合相关系数提高了4 %。说明RASTA滤波对信道噪声的低频成分中人耳听不到的部分进行过滤的方法在语音通信质量客观评价中是有效的。It is presented that the research result of using PASTA filter on objective communication speech quality evaluation.Perceptually Linear Prediction(PLP) and Mel Frequency Cepstral Coefficient(MFCC) methods with RASTA filter was used to compute the correlation between subjective and objective measure for the speech files which was transmitted in the shortwave and super shortwave channel.When RASTA filter is combined with PLP,the result is improved by 12%.When it is combined with MFCC,the result is improved by 4%.It proves that the RASTA filter which filters the lower frequency part of channel noise is useful in objective speech quality evaluation.

关 键 词：RASTA滤波 MFCC方法 PLP方法 语音音质客观评价 MOS值 

分 类 号：TN912.3[电子电信—通信与信息系统]