MR Noise is a broadband noise reduction processor that combines automatic noise profiling with an intuitive interface, offering real-time monitoring and frequency-dependent adjustments for clean, artifact-free audio.
Purpose and Core Operation
MR Noise is a frequency-domain processor designed to remove steady-state background noise such as hum, hiss, or room tone. It requires a noise profile, which it samples automatically when the Learn parameter is enabled, typically from leading silence in the audio. Once learned, the processor applies reduction based on that profile.
The default settings are tuned to work effectively in most scenarios, often producing clean results without manual tweaking. This makes it accessible for users who need quick cleanup, while still offering depth for more demanding restoration tasks.
Key Capabilities and Innovations
A standout feature is Auto Dynamics mode, which adjusts attack and release times in response to the input signal’s transient content. This preserves sharp percussive elements that conventional noise reduction might smear, while minimizing processing artifacts. The result is a balance between noise suppression and signal integrity.
The processor operates at low latency (under 20 milliseconds) and supports sample rates up to 192 kHz, enabling real-time use in live scenarios. Noise floor learning is also swift, down to 50 milliseconds, so lengthy noise loops are unnecessary for establishing a profile.
User Interface and Monitoring
The interface presents a comprehensive frequency response display showing the input spectrum, noise floor, and output spectrum simultaneously, allowing you to visually identify which frequencies are being attenuated. This real-time feedback simplifies adjustments and confirms processing effectiveness.
A dedicated monitor feature lets you audition only the removed noise, with its own volume control. This helps you hear exactly what is being subtracted, which is useful for fine-tuning reduction levels or diagnosing potential musical content loss.
Workflow and Practical Applications
Typical workflow begins by selecting a preset and pressing play; MR Noise learns the noise floor automatically and begins reduction. You can then adjust the Amount control for overall reduction intensity. For more precise control, you can edit the noise floor and reduction parameters across frequency bands by dragging control points directly on the display.
This makes MR Noise suitable for cleaning up dialogue, voiceovers, or musical recordings with consistent background noise. Its low latency and fast learning also allow for live performance or broadcast use where immediate processing is needed. Manual frequency editing supports mastering-level restoration where specific bands require different treatment.
Key features
- Frequency spectrum based noise reduction
- Low latency under 20 milliseconds for live use
- Fast noise floor learning down to 50 milliseconds
- Auto Dynamics mode preserves transients
- Monitor feature to hear removed noise
- Comprehensive display of input, output, and noise spectra
- Parameters editable as a function of frequency
Best for
MR Noise is best for engineers and sound designers seeking a fast, reliable noise reduction tool that handles typical background noise while preserving transients. Its low latency and quick learning make it ideal for live sound, dialogue post-production, and music cleanup, with manual frequency editing for fine-tuned restoration.