ScratchFree from Algorithmix is a DirectX/VST plug-in that can remove clicks, crackles and any other transient noises, with virtually no artifacts and no degradation of the original signal. Whether for preparing archive material for re-release on CD, SACD or DVD, or any other restoration project, the program has the ability to preserving timbre, ambience and low-level details in the original signal.
The software consists of two functions. A de-clicking module is used to remove severe clicks from old shellac and vinyl records, or switching noise originating from cross-talk or improper setups of digital audio equipment, while the de-crackling module removes any remaining small clicks and crackles. Both function are very simple to use with a minimum of controls.
A signal scope aids in finding ideal settings for the de-crackler process. Aside from restorative work, the plug-in can also aid in the masking of very small drop-outs in any audio source material.
- Effective removal of severe clicks from old shellac (78 RPM) and vinyl records
- Detection and removal of switching noise, static discharge, digital cross-talk, and thyristor buzz
- De-crackling algorithm preserving timbre, ambience, and low-level details in the original signal
- Flawless operation, suitable for high-resolution DSD post-production with support of up to 384kHz sampling frequency
- Can help with the reduction of distortion caused by signal clipping
- Presets simplifying restoration tasks in typical situations
- Clickless real-time parameter adjustment during playback
- Includes expert parameters for minimizing the appearance of artifacts, with virtually no artifacts generated when using correct settings for all parameters
- Signal scope for real-time tracking of the de-crackling process
- Difference function for real-time audio monitoring of the noise being removed
- Complete setup exchange between several simultaneously opened plug-ins
- Automatic latency compensation with compatible DAWs including Sadie, Wavelab, Cubase and Soundforge
- All internal calculations made using double floating-point accuracy (80-bits)