An Introduction to Audio Processing and Machine Learning Using Python

The pyAudioProcessing library classifies audio into different categories and genres. At a high level, any machine learning problem can be divided into three types of tasks: data tasks (data collection, data cleaning, and feature formation), training (building machine learning models using data features), and evaluation (assessing the model). Features, defined as ‘individual measurable propert[ies] or characteristic[s] of a phenomenon being observed,’ are very useful because they help a machine understand the data and classify it into categories or predict a value.

Source: opensource.com