Free and open source · MIT

Sonero Per-application audio mixer, router and equalizer for Linux.

Your game, your call, your music and your browser each get their own volume, their own equalizer and their own output device — live, without restarting anything. Built directly on PipeWire: no PulseAudio shims, no Wine, no vendor lock-in.

Needs PipeWire 0.3+ · x86-64 · the AppImage carries its own Qt

The Sonero mixer: one strip per channel with volume, mute, solo, balance and VU meters
Six output channels, each with its own volume, mute, solo, balance and live meters.

What it does

One mixer for every application you run

Sonero creates a set of virtual sound cards on PipeWire and lets you send each application to whichever one you like — then shapes each one separately.

Six output channels

System (master), Game, Chat, Media, Browser and Aux — volume, mute, solo, balance and live VU meters on each.

App routing that sticks

Every playing application shows up as a chip. Move it to a channel and it stays there across restarts.

Equalizer per channel

10, 15 or 31 bands with 24 built-in presets, plus save, import and export of your own curves.

Harmonic bass synthesis

A small driver cannot move enough air to reproduce 40 Hz. Sonero synthesises the harmonics of the bass in the range the driver does reproduce, and your hearing supplies the fundamental that is not there.

Themes you can write

Thirteen colours and a font in a JSON file; the shades in between are derived from them. The built-in theme is compiled in and only ever applied over, so nothing you write can leave the app unusable.

Automatic gain

On request, it holds the highest gain — up to +6 dB — that keeps the channel clear of full scale. Slow up, immediate down, so nothing breathes.

Every output, equally

USB, Bluetooth, HDMI, S/PDIF and onboard analog are all detected and routable — and two channels can play to two different devices at once.

Battery, without a product list

Read from the kernel’s own power_supply class, which covers any device whose driver reports a level, plus BlueZ for Bluetooth. A vendor protocol is one more source behind the same interface, not the feature itself — shown in the tray either way.

Streaming mix for OBS

A Sonero Stream source with an independent send level per channel, so your viewers hear a different balance than you do.

Virtual microphone

Partial

A Sonero Microphone source your apps can select like any other. Recording works today; the gain, mute and level controls are laid out but not yet wired, and the app greys them out.

A real desktop app

Tray icon with device battery and a compact volume window, notifications, autostart at login, and background operation so channels stay alive.

Equalizer

It runs in software, so it works with any headset

Every channel is a cascade of peaking biquads — nothing has to be blessed by a hardware vendor. Drag the curve or the per-band sliders; 10, 15 or 31 bands, with presets you can save and share. The harmonic bass stage sits right after it, on the same channel.

The order is not a preference: the equalizer feeds the bass stage so its crossover sees the signal as it will be heard, a brick-wall limiter follows both because it is the only thing between them and a driver, and the fader comes last so pulling a channel down never changes what the limiter was protecting against. Ahead of the filters the signal is attenuated by exactly what the curve boosts, so a raised band cannot clip on the way out.

The per-channel equalizer: a draggable response curve above per-band sliders

Themes

The interface is a file you can edit

Thirteen colours and a font in a JSON file. The shades in between — a hover against its surface, a slider groove, the tint behind a soloed channel — are derived from those thirteen, so a coherent interface costs thirteen decisions rather than thirty-six.

The built-in theme is compiled into the binary and is only ever applied over, never replaced, so a theme that is wrong or written for an older version can change colours and nothing else. Import, export and switch from the Profiles page; an optional style-sheet block covers what colours cannot say.

The Profiles page: a theme picker with import and export

Microphone

A voice chain, not just a level

The microphone Sonero hands to your applications runs a full broadcast chain: a high-pass, learned noise suppression, a gate that fades between sentences instead of chopping them, a de-esser, a ten-band voice equalizer, a compressor and a brick-wall limiter.

The noise model is RNNoise, and it runs ahead of everything else so the gate and the compressor measure speech against silence rather than against the room. Presets for podcasting, streaming and meetings; Raw switches every stage off.

The Sonero microphone page: input gain and level, the voice equalizer curve, and noise suppression

Install

Pick the file for your distribution

Every release carries all of these. Pushes to main also refresh a dev prerelease, if you would rather run what is being worked on. On Arch there is nothing to download — build the package from the repository instead.

FileFor
Sonero-<version>-x86_64.AppImage Take this one. Portable, carries its own Qt, built on the oldest glibc so it runs almost anywhere.
sonero_<version>-ubuntu2204_amd64.deb Ubuntu 22.04, Linux Mint 21.x
sonero_<version>-ubuntu2404_amd64.deb Ubuntu 24.04, Linux Mint 22.x
sonero_<version>-ubuntu2604_amd64.deb Ubuntu 26.04, Linux Mint 23.x
Sonero-<version>-arch-x86_64.AppImage Arch, if you would rather not build — otherwise the package below
Sonero-<version>-fedora-x86_64.AppImage Fedora, or a distribution at least as new
AppImage
# needs FUSE 2, or --appimage-extract-and-run
chmod +x Sonero-*-x86_64.AppImage
./Sonero-*-x86_64.AppImage
Debian · Ubuntu · Mint
# apt pulls in Qt and installs the udev rule
sudo apt install ./sonero_*-ubuntu2204_amd64.deb
Arch · Manjaro · EndeavourOS
# builds a real package and installs it with pacman
git clone https://github.com/Tatsu21/sonero.git
yay -B sonero/packaging/aur/sonero
From source
# the nicer route on Fedora and anywhere without a package
cmake -S . -B build -G Ninja \
      -DCMAKE_BUILD_TYPE=Release
cmake --build build --parallel
./build/Sonero

The AppImage adds itself to your application menu on first run — and deletes the bundle it replaces, so old versions do not pile up in your downloads folder. On GNOME the tray icon needs the AppIndicator extension.

Honest status

What is not finished

These are visible in the app but greyed out and labelled, rather than left looking clickable. Nothing here pretends to work.

  • Profiles — the page exists and holds themes, but saving a whole audio configuration as a named profile is not implemented.
  • Microphone gate and monitoring — the source is live, and input gain, mute, the voice equalizer and noise suppression all reach it. The noise gate and “hear myself” are laid out but not wired; the DSP for them exists and is tested, and has no node to run in yet.
  • Noise suppression needs RNNoise at build time. The released .deb and AppImage carry it. A build from source finds it only on Arch, or after packaging/build-rnnoise.sh — without it that one stage reports itself unavailable and says so in the app.
  • SteelSeries hardware EQ — the dock accepts the HID writes and nothing audible changes. Unsolved, so the card is disabled; the per-channel equalizer runs in software and works with any headset.
  • LDAC is untested. The codec picker ranks it highest, but none of the headsets on hand offers it, so that path has never run on real hardware.

Requirements

Nothing exotic

ComponentMinimumNotes
PipeWire0.3 Plus WirePlumber as the session manager. Already running on every current desktop.
Qt6.2 Widgets, D-Bus and Network. The AppImage carries its own.
CPUx86-64 The DSP is a handful of biquads per channel.