欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对不同编码算法对语音质量的影响，采用语音质量改进参数构造特征矢量，提出了一种基于SVM多级判决的语音编码检测算法，实现了对不同信道的识别。采用统计方法对这组参数的区分性进行了分析，在此基础上设计了一种高效的信道检测方案，结合实际数据，测试了算法性能，并分析了语音长度对性能的影响。实验结果表明，该算法能有效地提高信道检测准确率。
英文摘要：
According to the different speech quality distortion caused by different coding algorithms, a speech transmission channel detection method based on SVM multi-stage decision is proposed using the feature vectors made of modified speech quality parameters. After analyzing the discriminability of the parameters using statistical approach, a high-efficient channel detection scheme is designed. The performance of the method is tested based on speech collected from these channels, and the influence by speech length on the performance is also discussed. Experimental results show that the algorithm can improve channel detection accuracy effectively.
同期刊论文项目
同项目期刊论文
期刊信息
位置：