Awesome Linux Game Tools is a series of reviews showcasing the finest tools for Linux gamers.
Noise suppression is a pretty old topic in speech processing, dating back to at least the 1970s. As the name implies, the idea is to take a noisy signal and remove as much noise as possible while causing minimum distortion to the speech of interest.
noise-suppression-for-voice is a noise suppression plugin based on RNNoise, a noise suppression library based on a recurrent neural network (RNN). A RNN is a class of artificial neural networks where connections between nodes can create a cycle, allowing output from some nodes to affect subsequent input to the same nodes. RNNs are particularly effective for background noise removal because they can learn patterns across time which is essential for understanding audio.
noise-suppression-for-voice isn’t only useful for gamers who are live streaming and recording. It can be used to suppress noise in a wide range of applications.
Installation
We tested the software in Arch. There’s a package in the Arch User Repository, which we installed with the yay helper.
$ yay -S noise-suppression-for-voice
There’s manual configuration needed. We need to create a config directory
$ mkdir ~/.config/pipewire/
Next create a pipewire.conf.d directory.
$ mkdir ~/.config/pipewire/pipewire.conf.d/
We need to create a configuration file (99-input-denoising.conf) for the plugin. Use any text editor such as nano.
$ nano ~/.config/pipewire/pipewire.conf.d/99-input-denoising.conf
Paste the contents below into that file.
context.modules = [
{ name = libpipewire-module-filter-chain
args = {
node.description = "Noise Canceling source"
media.name = "Noise Canceling source"
filter.graph = {
nodes = [
{
type = ladspa
name = rnnoise
plugin = /usr/lib/ladspa/librnnoise_ladspa.so
label = noise_suppressor_mono
control = {
"VAD Threshold (%)" = 50.0
"VAD Grace Period (ms)" = 200
"Retroactive VAD Grace (ms)" = 0
}
}
]
}
capture.props = {
node.name = "capture.rnnoise_source"
node.passive = true
audio.rate = 48000
}
playback.props = {
node.name = "rnnoise_source"
media.class = Audio/Source
audio.rate = 48000
}
}
}
]
You need to specify where librnnoise_ladspa.so is stored. With the AUR package that’s at /usr/lib/. We’ve bolded the line which you need to edit if the .so file is stored in a different location on your system.
Restart PipeWire with the command:
$ systemctl restart --user pipewire.service
If you’re running PulseAudio instead of PipeWire, you’ll need to follow different configuration steps. They are detailed on the project’s GitHub page. We only tested noise-suppression-for-voice with PipeWire.
Next page: Page 2 – In Operation and Summary
Pages in this article:
Page 1 – Introduction and Installation
Page 2 – In Operation and Summary