典型的音频分类算法

Typical Methods for Audio Classification

音频分类是提取音频结构和内容语义的重要手段,是基于内容的音频检索和分析的基础。本文对几种常用的音频分类算法作了综述,介绍了最小距离法、神经网络、支持向量机、决策树方法、隐马尔可夫模型等典型算法的特征,并对它们的优缺点进行了比较。

Audio classification is an important method to extract audio structure and content, and is a basis for further audio retrieval and analysis. This article gives an overview of existing techniques for audio classification such as minimum distance classitier, neural network, support vector machines, decision tree, and hidden Markov Modal etc. And their merits and limitations are also contrasted.