Summary
Noise suppression for voice is a real-time noise suppression plugin based on Xiph's RNNoise library. It is designed to reduce various background noises in voice recordings or streams while preserving voice quality as much as possible.
Features
- Supports multiple plugin formats: VST2, VST3, LV2, LADSPA, AU, AUv3.
- Operates only with 48 kHz sample rate and 16-bit audio input.
- Designed to suppress noise from sources such as computer fans, office environments, crowds, airplanes, cars, trains, and construction.
- Has a minimalistic graphical user interface (GUI) to adjust parameters and view diagnostic statistics.
- Configurable parameters include Voice Activity Detection (VAD) Threshold, VAD Grace Period, and Retroactive VAD Grace Period to fine-tune noise suppression behavior.
- Includes detailed instructions for use on Windows with Equalizer APO and on Linux with PipeWire and PulseAudio.
- Open source under GPL-3.0 license and actively maintained on GitHub.
Limitations
The plugin may not improve voice quality if the microphone itself is poor; it may even degrade it by misclassifying voice as noise.
Platforms
Explicitly supports Windows and Linux platforms through mentioned middleware (Equalizer APO on Windows, PipeWire and PulseAudio on Linux). No other platforms are indicated.
License and Availability
The software is licensed under GPL-3.0 and is available free of charge as open source software on GitHub.
Comments
Sign in to rate, comment and bookmark.
No comments on this tool yet.