Matchering

media

Open source audio mastering tool that matches the loudness, spectrum, and stereo width of your track to a reference track. Self-hostable alternative to paid mastering services like LANDR and eMastered

#audio#mastering#music#dsp#self-hosted#python

Quick Start

pip install matchering && python -c "import matchering as mg; mg.process('my_mix.wav', 'reference.wav', 'output.wav')"

Overview

Matchering is a Python library and CLI tool for automated audio mastering based on reference matching. You provide a mix file and a commercially released reference track in the same genre, and Matchering analyses the reference’s loudness, frequency spectrum, and stereo width before applying processing to bring your track to a similar profile.

The approach is the same technique professional mastering engineers use when clients send a reference track alongside their mix. Rather than applying a fixed loudness target or a static EQ curve, Matchering derives its processing chain from the specific characteristics of the reference you choose. The result is a master calibrated to a real-world commercial standard rather than an arbitrary algorithm’s idea of “loud.”

Batch processing is where the tool earns its value for working musicians and producers. The Python API makes it straightforward to loop through an album’s worth of tracks, master each one against the same reference, and produce a consistent loudness and spectral balance across the release. Services like LANDR charge per track; Matchering runs locally with no per-use cost once deployed.

A companion project, matchering-web, adds a browser-based interface for users who prefer not to work from the command line. It deploys as a separate application and connects to the core Matchering library for processing.

The limitation is control. Matchering makes all mastering decisions automatically based on the reference. You cannot dial in multiband compression ratios, manually adjust the high shelf, or make the kind of creative mastering choices a human engineer would. For tracks that need precise creative control, it is a starting point or a rough approval master rather than a final deliverable.

Use Cases

Specific ways to use Matchering for your workflow.

01
Master a finished mix to match the loudness and tonal balance of a commercial reference track without paying per-track fees
02
Batch-process album tracks to achieve consistent loudness and spectral balance across the release
03
Prototype a master for client approval before committing to a paid mastering session or engineer
04
Integrate automated mastering into a music production pipeline or DAW export workflow via the Python API

Deployment Strategy

Recommended ways to host Matchering in your own environment.

docker
self-hosted