摘要
本文使用支持向量机自动对体育节目中的音频精彩场景识别。由试验结果可以看出 ,在小样本前提下 ,采用支持向量机能够保证训练误差和实际误差均小 。
Audio information exists at everywhere, but is often out-of-order. It is necessary to arrangy them into regularized classes and recognize in order to use them more easily. In this paper, we present our work in applying support vector machines(SVMs) to recoginze audio highlight scene in sports program. In our experiment, Useings VSMs do well in train error and result error if we select a small sample, so it have a high precision.
出处
《微计算机应用》
2004年第3期316-320,共5页
Microcomputer Applications
