vst-bench-ml: Auditable Dataset Generation for Audio Plugins
Abstract
Python software for sampling audio-plugin parameters, coordinating rendering through a separately installed localhost host, retaining records of outputs and failed/dropped attempts, and checking sealed dataset integrity offline. This source snapshot excludes private research modules and contains no plugin binaries, datasets, audio, presets, or model weights. The companion host is archived separately. Demonstrated repeatability is limited to the documented macOS arm64/Python 3.12 setup; no cross-platform or scientific-validity claim.
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 23 September 2026