大家好，我是小鸭酱，博客地址为：http://www.cnblogs.com/xiaoyajiang
pyAudioAnalysis是一个音频分析python库，用于Feature Extraction, Classification, Segmentation 和Applications，其github见 https://github.com/tyiannak/pyAudioAnalysis
由于时间紧凑，每次更新一点，请见谅。
1 安装
非常容易，如果你使用的是Linux下的pycharm，那么new一个project，再设置其解释器为python2.7，然后再点击绿色的+号，搜索pyAudioAnalysis，点击install，那么pycharm就可以自动为你解决好所有的依赖问题。好开心。当然，你如果不想用pycharm，你可以直接使用pip安装，见https://github.com/tyiannak/pyAudioAnalysis/wiki/2.-General，也很容易。
2 功能介绍
安装好以后，可以看到其包含如下6个python文件：
| python文件名 | 功能概览 |
| audioAnalysis.py | this file implements the command-line interface of the basic functionalities of the library, along with some recording functionalities. |
| audioFeatureExtraction.py | this is where all audio feature extraction is implemented. In total, 21 short-term features are computed, while a mid-term windowing technique is also implemented, in order to extract statistics of audio features. |
| audioTrainTest.py | this file implements the audio classification prodecures. It contains functions that can be used to train a Support Vector Machine or k-Nearest-Neighbour classifier. Also, wrapper functions and scripts are provided for general training, evaluating and feature normalization issues. |
| audioSegmentation.py | this file implements audio segmentation functionalities, e.g. fixed-sized segment classification and segmentation, speaker diarization, etc. |
| audioBasicIO.py | this file implements some basic audio IO functionalities as well as file convertions |
| audioVisualization.py | the purpose of this set of functions is to produce user-friendly and representative content visualizations |
In the data/ folder, a couple of audio sample files are provided, along with some trained SVM and kNN models for particular classification tasks (e.g. Speech vs Music, Musical Genre Classification, etc).
2.1 特征提取(Feature Extraction)
2.2 分类和回归(Classification and Regression)
2.3 分割(Segmentation)
2.4 数据可视化(Data-visualization)
2.5 音频再识别功能(Audio-Recording-Functionalities)
2.6 其他功能