摘要
采用能够反映人对语音的感知特性的线性预测(LPC)倒谱参数作为特征参数,同时针对支持向量机技术对模式识别中的非线性、高维数的样本问题有非常好的分类效果和学习推广能力,设计了一个支持向量机分类器来进行说话人识别。试验结果验证了该系统有很高的识别率和较强的鲁棒性。
This paper presents a speaker recognition system which uses LPCCEP parameters,and devises an Support Vector Machines(SVM)classify system to have speaker recognition as SVM is a classifier devised for nonlinear and high dimensional sample problems and has an effective classify and learning ability.The system based on that has obtained better recognition rate and robustness.[
出处
《电声技术》
北大核心
2004年第3期39-41,44,共4页
Audio Engineering
