While investigating how services like Shazam worked, how they identified music using audio samples, I stumbled upon Echoprint, an open-source music identification project, and when I did, I immediately jumped to experiment with it.

The Echoprint project has many sub-projects in itself, but the simplest way to use it is to send the audio sample in a specific format to an Echoprint server. The Echoprint project already has a free-to-use (keep in mind that with “free” comes request limitations) online Echoprint server that identifies the music for you, so you only need to create an appropriate code to call the service.

The echoprint-codegen project is used exactly for that purpose: it takes an audio input and produces a code out of it, ready to use in Echoprint service requests. There are no binary distributions for the project, so you need to manually build it after downloading (or cloning) the project on your PC.

First, clone the project from the Github repository.