www.kvraudio.com/news/hornet-plugins-releases-zerocomp---ai-compressor-with-three-models-64959
7th October 2025
HoRNet Plugins has announced the release of ZeroComp, an intelligent compressor that automatically learns audio material characteristics and self-configures for optimal compression results.
The plugin features three compression models and proprietary crest factor analysis, offering what HoRNet says is a new approach to automated dynamics processing.
ZeroComp differentiates itself from existing AI-powered compressors through its three-model processing system. Unlike competitors that rely on single-algorithm approaches, ZeroComp simultaneously runs three distinct compression models:
Users can switch between models instantaneously during playback without audio interruption, enabling real-time A/B comparisons and creative flexibility.
The plugin's intelligence stems from crest factor analysis that examines the ratio between peak and average signal levels. This data drives automatic calibration of all critical compression parameters including threshold, ratio, attack, release, knee, and makeup gain. The system's convergence technology ensures stable, oscillation-free processing while providing visual feedback on learning progress.
The plugin's operation reflects HoRNet's philosophy of accessible professional tools. Users simply insert ZeroComp on the desired track, initiate playback for analysis, and monitor convergence indicators as the plugin creates optimized compression models. Once learning is complete, the system provides immediate professional-quality processing.
The interface features hardware-accelerated graphics with HoRNet's minimalist design approach, optimized for both standard and high-resolution displays.
ZeroComp is available now for €24,99 for macOS 10.13+ and Windows 10+ operating systems with native Apple Silicon optimization. The plugin is available in Audio Units (AU), VST3 and AAX formats, ensuring compatibility across all major digital audio workstations.
KVR Audio, Inc.
www.kvraudio.com