PipeTune

Engine and User Interface for Applied EffeTune DSP on a Linux Desktop

---

(Japanese language is here/日本語はこちら)

Please note that this English version of the document was machine-translated and then partially edited, so it may contain inaccuracies. We welcome pull requests to correct any errors in the text.

What Is This?

PipeTune applies an EffeTune DSP preset to all audio in one Linux desktop session.

EffeTune DSP performs computations using fully natively compiled binaries. On each platform, you can select native SIMD operations.

WirePlumber (PipeWire orchestrator) inserts PipeTune as a transparent filter between the mixed desktop playback stream and the normal PipeWire output path. A GTK3 control application remains available through the desktop system tray.

Features

• You can load EffeTune preset files to apply a DSP pipeline to the audio output of the entire Linux system.
• Loads standard and legacy EffeTune preset files with the .effetune_preset extension and applies the DSP pipeline to desktop audio.
• Supports the EffeTune 2.7.0 native DSP contract, including direct runs with 1 through 16 planar channels.
• Automatically negotiates the sampling rate with the PipeWire graph, or computes the DSP at specified rates of 44.1, 48, 96, 192, or 384 kHz.
• The DSP performs computations entirely in native code. You can choose between Scalar (for compatibility), automatic SIMD selection, or CPU-verified implementations for specific instruction sets.
• Automatically suspends DSP work after a selectable period of silent input while allowing effect tails to finish first.
• A GTK application that resides in the system tray allows you to manage various DSP states and settings.
• Operation is also possible via CLI commands.

Supported systems

PipeTune requires a PipeWire desktop session managed by WirePlumber and systemd user services. WirePlumber 0.4 and 0.5 are supported. A standalone

Note: This applies to the standard Debian and Ubuntu distributions. It may also work on other distributions if they meet the system requirements.

Prebuilt Debian packages are published for:

Distribution	Release	Architectures
Debian	bookworm	amd64, i386, arm64, armhf
Debian	trixie	amd64, i386, arm64, armhf, riscv64
Ubuntu	24.04	amd64, arm64
Ubuntu	26.04	amd64, arm64

---

Download and install

Download the .deb matching your distribution, release, and architecture from the PipeTune GitHub Releases page.

Use the following command if you are unsure which package architecture to download:

dpkg --print-architecture

Open a terminal in the directory containing the downloaded package. Make sure that directory contains only the intended PipeTune package, then install it with:

sudo apt install ./pipetune-*.deb

apt installs the package and its required runtime dependencies.

To verify that the installation succeeded, run:

pipetune --version

Initial setup

Since PipeTune runs as a user-level daemon, setup is required once for each desktop user after the package is installed. Double-clicking the PipeTune icon in the GNOME application launcher, or launching it from another desktop application menu, performs this setup automatically before the GTK application connects to the daemon. This also covers a user launching PipeTune for the first time when the package was installed earlier by someone else.

To perform the same setup explicitly, run this as the regular desktop user, without sudo:

pipetune setup

setup checks the current user's managed files, setup version, and systemd user service. If the integration is already current, it exits without repeating setup. Otherwise it installs or repairs the integration and launches PipeTune GTK in the system tray. Use pipetune setup --force (or the short form pipetune setup -f) to repeat the setup operations unconditionally.

During an automatic first launch, the window displays setup progress and keeps its controls read-only. If setup fails, PipeTune GTK opens the Action Log with the diagnostic and sends a desktop notification when the window is hidden. It still attempts to connect so an already-running daemon remains usable.

Open the PipeTune settings window by double-clicking the system tray icon or selecting Open from its menu.

PipeTune settings window

The PipeTune settings window always displays PipeTune's status, divided into sections, on the left, while the Processing, Rate, DSP, and Advanced settings are shown on the right.

Changing a setting previews it immediately in the running daemon. The Apply button confirms and saves the current state. The Cancel button, Escape, or the title-bar close button restores the previous live state before hiding the window.

The status pane on the left displays DSP utilization as a bar. The Action Log drawer at the bottom lets you review the history of connections, previews, saves, and failures.

Use the Enable DSP processing switch under Processing to easily turn DSP processing on or off (bypass mode).

The DSP page can also suspend DSP work automatically after continuous silent input. Its duration control accepts 0.1 through 5.0 seconds in 0.1 second steps.

If the PipeTune daemon cannot be reached, the UI controls become read-only and normal operation resumes after reconnection.

Audio streams and PipeTune

PipeTune uses PipeWire and WirePlumber on Linux to insert DSP processing into the audio stream as a filter, as shown below:

flowchart LR
    apps1["Application (1)"]
    apps2["Application (2)"]
    apps3["Application (3)"]
    mix["PipeWire mix"]
    tune["PipeTune filter<br/>EffeTune DSP or bypass"]
    output["PipeWire output<br/>master volume and default route"]
    device["Selected audio device"]
    device2["Audio device (2)"]
    device3["Audio device (3)"]

    apps1 --> mix
    apps2 --> mix
    apps3 --> mix
    mix --> tune
    tune --> output
    output --> device2
    output --> device
    output --> device3

Loading

Per-application volume is applied before mixing. In bypass mode, the mixed PCM is passed downstream without invoking EffeTune. After the DSP filter, desktop master-volume and output-device selection work exactly as they do without PipeTune.

Selecting an EffeTune DSP preset

When loading an EffeTune DSP preset into PipeTune, you can select from:

• EffeTune's standard DSP presets
• User presets saved by EffeTune for Linux AppImage
• An individual *.effetune_preset file

The user preset file saved by the Linux AppImage is located at $XDG_CONFIG_HOME/effetune/effetune_presets.json (or ~/.config/effetune/effetune_presets.json).

Select a preset from the list or specify a file to preview it immediately in the DSP.

While preset processing is active, PipeTune monitors the file used by the DSP. An in-place edit or atomic replacement is loaded automatically. The new pipeline is built completely before activation; if reading, parsing, or DSP construction fails, the currently running pipeline remains active and the error is shown in runtime status. A later valid update is retried automatically. When the active choice is a preset saved by EffeTune, the settings application refreshes its private standalone snapshot after the corresponding entry changes, which triggers the same daemon-side reload.

From the CLI, you can specify either a preset file path or bypass mode:

pipetune setup --preset /absolute/path/to/example.effetune_preset
pipetune bypass

Choosing the PCM rate

Configure the rate using DSP sampling frequency and PipeWire enforcement in the PipeTune settings window. Automatic follows the rate negotiated by the PipeWire graph. A fixed setting requests 44.1, 48, 96, 192, or 384 kHz for both filter nodes and the EffeTune engine. The status display shows the actual DSP and graph rates.

• Suggest proposes the graph rate to PipeWire. PipeWire may accept it or choose a different graph rate to accommodate the audio streams.
• Force additionally asks PipeWire to maintain that rate.

Neither mode changes PipeWire's global clock configuration.

If the requested value cannot be applied, PipeTune remains connected and continues DSP processing at the graph rate that was actually negotiated. An unsuccessful Force request appears in the Rate diagnostics.

The same information and controls are available from the CLI:

pipetune rate list
pipetune rate get
pipetune rate set automatic
pipetune rate set 192000 force

rate list lists Automatic and the five fixed rates. rate get reports the configured policy and negotiated rates. While connected to the daemon, rate set applies the change immediately and saves it only after the daemon confirms success. If the daemon is unavailable, the setting is saved for its next start.

Changing the rate may cause a brief period of silence while PipeTune rebuilds the DSP and PipeWire streams.

Choosing the native DSP backend

PipeTune lets you choose which CPU instruction set is used to compute the EffeTune DSP. Select it from Native backend on the DSP page of the PipeTune settings window.

• Scalar is the compatibility-oriented default. It is intended to reproduce EffeTune's computations as faithfully as possible.
• SIMD (Auto) automatically selects the highest validated implementation supported by the CPU.
• The same drop-down can also pin baseline, x86-64-v3, x86-64-v4, or Arm64 SVE where available on the target architecture.

Note: The original EffeTune performs its computations using WebAssembly and WebAssembly SIMD, whereas PipeTune uses native CPU instructions. Therefore, very small numerical differences from EffeTune may occur. The project owner could not hear any difference.

The same information and controls are available from the CLI:

pipetune dsp list
pipetune dsp get
pipetune dsp set scalar
pipetune dsp set simd
pipetune dsp set simd --variant x86-64-v3

While connected to the daemon, dsp set rebuilds the current preset pipeline with the new backend and saves the setting only after the daemon confirms success.

DSP internal state is reset, so a discontinuity or brief silence during the switch is expected. If the daemon is unavailable, a locally validated choice is saved for its next start. If the configured SIMD backend fails its CPU requirements or library validation at startup, PipeTune falls back to an available lower SIMD tier or Scalar and reports the actual variant and reason in the status display.

Suspending DSP during silence

Turn on Suspend DSP on silence on the DSP page to avoid continuing EffeTune calculations when an application, such as a paused browser video, keeps sending silent PCM. Silence duration selects 0.1 through 5.0 seconds in 0.1 second steps. Turning the switch off selects ignore, which is the default and keeps processing silent input as before. The last selected duration remains visible while the switch is off and is reused when it is turned on again; the first selection is 1.0 second.

PipeTune checks the PCM after input sample-rate conversion. During the selected silent interval it continues running the complete preset, so delay, reverb, and generated effect output can finish. It then fades the result to zero over 5 ms, resets the DSP state, emits zero PCM, and stops invoking EffeTune. The first block containing any nonzero sample wakes the DSP and is processed immediately. Effects that generate sound from otherwise silent input, such as vinyl-noise simulations, are also stopped after the selected interval when this option is enabled.

The status pane keeps the preset processing mode unchanged and reports DSP activity separately. While the silent interval is being processed it shows Draining; after DSP work stops, the Load display shows Suspended instead of a percentage.

The saved environment assignment is PIPETUNE_DSP_IDLE_TIMEOUT=ignore or an integer number of milliseconds from 100 through 5000 in 100 ms steps.

Resetting PipeTune configuration

Restore Defaults on the Advanced page of the PipeTune settings window previews the following state without saving it:

• DSP changes to bypass mode
• PCM rate is Automatic and the PipeWire request is Suggest
• Native DSP backend is Scalar
• Automatic DSP suspension is Ignore

Use Apply to save the defaults or Cancel to restore the previous live settings. This GTK operation does not restart the service.

You can also reset the configuration from the CLI:

pipetune config reset
pipetune config reset --yes

Without --yes (or -y), the CLI asks for confirmation. The configuration file is completely replaced, so this command can also recover from a corrupted configuration file.

A running user service is restarted immediately; a stopped service remains stopped.

Update or remove PipeTune

To update PipeTune, download a newer matching package from GitHub Releases and install it with the same sudo apt install ./pipetune-*.deb command.

Then launch PipeTune from the desktop application menu. It detects the updated setup version and applies any required per-user changes. The equivalent manual command is pipetune setup as the regular desktop user.

Before removing the package, undo its per-user integration as the desktop user:

pipetune unsetup
sudo apt remove pipetune

pipetune unsetup quits the GTK application, disables and stops the user service, and removes PipeTune's WirePlumber configuration. It also creates a per-user autostart mask to prevent the GTK application from starting automatically. The startup preset selection is preserved.

Use pipetune unsetup --purge to also remove PipeTune's application configuration.

Logs

View daemon logs with:

journalctl --user -u pipetune.service

---

More information

• Daemon operation and developer documentation
• GTK application behavior
• Native DSP backends and benchmarking

Limitations

FIR Crossover, 5Band FIR PEQ, Group Delay EQ, and Group Delay PEQ are supported. PipeTune regenerates their convolution coefficients from the preset parameters and the active sample rate. FIR Crossover requires an even output bus from 4 through 16 channels and is omitted with a warning on other layouts.

Room EQ and IR Reverb remain unsupported. A Room EQ preset references measurement data that EffeTune resolves through its measurement store, and IR Reverb resolves a content identifier through its IR library. The required PCM is not contained in .effetune_preset, so PipeTune omits these nodes with warnings.

License

Under MIT.