Abstract: Feature extraction is the foundation of audio classification，while audio classification is a key technology of content based audio retrieval.In this paper，discriminating features between speech and music are analyzed，the work on audio feature extraction and classification based on wavelet transform and Support Vector Machine（SVM） is presented.It is used to classify audio into pure speech，music，speech with music and environment sounds.The performance of some new proposed feature is evaluated.The experiments results show that the feature selected are effective for audio classification，and the classification accuracy is good.

Key words: wavelet transform, feature extraction, audio classification, Support Vector Machine（SVM）

摘要： 音频特征提取是音频分类的基础，而音频分类又是内容的音频检索的关键。综合分析了语音和音乐的区别性特征，提出一种基于小波变换和支持向量机的音频特征提取和分类的方法，用于纯语音、音乐、带背景音乐的语音以及环境音的分类，并且评估了新特征集合在SVM分类器上的分类效果。实验结果表明，提出的音频特征有效、合理，分类性能较好。

关键词: 小波变换, 特征提取, 音频分类, 支持向量机