AI Mastering

The engine analyzes the incoming mix and applies compression, equalization, and limiting in one pass. You get a louder track with the frequency balance pulled toward commercial references, without touching a single plug-in. The catch is that the decisions are the algorithm's, not yours.

You set the loudness you want, expressed in decibels, and the limiter pushes the track toward that figure. Wanting more volume means raising the target, as the official documentation puts it. This makes it easy to hit a platform's preferred level. Push too far and you lose dynamic range.

A separate control sets how aggressively the master is processed. Lower settings keep more of the original dynamics, higher settings flatten the track toward a loud, dense sound. For acoustic, jazz, or classical work, a gentler setting usually sounds closer to the source.

The interface shows a spectrum view and loudness readings for the processed file. You can see which frequency bands the master moved and how far the overall level shifted, which helps you judge whether the result suits the genre.

An option reduces low frequency attenuation during processing. Dance, hip hop, and electronic producers use it to keep the kick and sub bass from thinning out after limiting, a common complaint with automatic mastering.

The same algorithm ships as phaselimiter, an open source C++ program, with a GUI build for Windows. Running it locally removes the online queue and the file size ceiling, limited only by your computer's memory. Linux and macOS users have to build it from source.

A Go based command line client masters files from a terminal with a single command, and no account setup is needed because each run generates a guest API key. The cloud service also exposes an OpenAPI specification for teams that want to integrate mastering into their own pipeline.