Matchering
Matches a target track to a reference for producers, engineers, and developers
Matchering takes a TARGET audio file and a REFERENCE audio file, then creates a mastered version of the target that follows the reference’s RMS, frequency response, peak amplitude, and stereo width. Optional FFmpeg support adds MP3 loading, while its Hyrax brickwall limiter handles limiting. The Python 3 rewrite no longer depends on MATLAB and improves processing speed and accuracy compared with the previous major version. It is available as a containerized web application, a Python library, and a ComfyUI node, and it is also integrated into the UVR5 Desktop App. Docker Compose is listed as an installation method, though the research does not provide the compose file or detailed setup steps.
The program is intended for music producers and audio engineers who want reference-based matching, as well as AI mastering startups and developers using its available interfaces. Access options include desktop, node-based, library, and web application forms.
PaidProduct is named as the paid product it replaces. Its overlap with Matchering and any feature gaps are marked unknown.
Key features
- Matches reference RMS, frequency response, peak amplitude, and stereo width
- Runs as a web application, Python library, or ComfyUI node
- Integrates with the UVR5 Desktop App
- Rewritten in Python 3 without MATLAB
- Includes the Hyrax brickwall limiter
- Improves speed and accuracy over the previous major version
- Loads MP3 files with optional FFmpeg support
vs PaidProduct
Matchering is listed as a replacement for PaidProduct. It masters a target from a reference using RMS, frequency response, peak amplitude, and stereo width, but the documented overlap and gaps with PaidProduct are unknown. It is intended for music producers, audio engineers, AI mastering startups, and developers seeking this reference-based workflow.