Manual mix
This version was finished by manually refining each stage. It serves as a reference for ShushuLab’s human mixing decisions.
Listen to two versions of the same song, “D.O.A”, switching instantly at the same playback position.
This is not a claim that semi-automated mixing is better. These examples show the results of a production process that moves repetitive processing into tools while retaining human judgment.
Press Play to load the comparison audio. It loops smoothly for the duration of the supplied files while you switch between the mixes.
Listen to A and B separately with your browser's native audio controls. This fallback does not provide synchronized switching or level matching.
The original stereo mixes differ in loudness by about 0.6 LU. By default, B is lowered by about 0.6 dB to reduce the influence of that difference. Uncheck the option to listen with the original relative levels.
Both web listening files use the same headroom processing and encoding settings.
For listening loops up to 30 seconds, the A/B playback boundaries use the same approximately 40 ms crossfade to reduce gaps and clicks. The published source files are not modified.
This version was finished by manually refining each stage. It serves as a reference for ShushuLab’s human mixing decisions.
In this version, repetitive processing and stages that can be standardized were handled by ShushuLab’s own plugins, with final decisions made by a person. It is not a fully automated mix.


The photographs document the production and performance background. No specific venue, date, equipment or production credit is inferred from the photographs.
The audio is provided for production comparison and listening. Unauthorized reposting, redistribution or secondary use of the audio files is prohibited.