Matchering is an open-source audio matching and mastering toolkit that masters a target track by analyzing and matching it to a reference track. It adjusts key perceptual and technical properties so the processed target more closely resembles the reference.

Key Features

Reference-based mastering workflow using a target and a reference audio file
Matches loudness (RMS), frequency response (tonal balance), peak level, and stereo width
Brickwall limiting via the built-in Hyrax limiter
Usable as a Python library for integration into pipelines and products
Containerized deployment option (Docker image) for running as a web application

Use Cases

Quickly create a consistent sound across multiple songs in an album or EP
Produce alternate masters that emulate different commercial references
Integrate automated reference mastering into developer workflows or services

Limitations and Considerations

Quality depends heavily on selecting an appropriate reference track and starting mix
Some audio formats (for example MP3) may require FFmpeg to be installed
Requires a reasonably capable machine (the project recommends at least 4 GB RAM)

Matchering is best suited for creators and developers who want repeatable, reference-driven mastering with an open implementation. It can be used interactively or embedded into automated processing workflows.