A human readable description of the node or stream.
@PAR@ node-prop media.name
A user readable media name, usually the artist and title.
These are usually shown in user facing applications
to inform the user about the current playing media.
@PAR@ node-prop media.title
A user readable stream title.
@PAR@ node-prop media.artist
A user readable stream artist
@PAR@ node-prop media.copyright
User readable stream copyright information
@PAR@ node-prop media.software
User readable stream generator software information
@PAR@ node-prop media.language
Stream language in POSIX format. Ex: `en_GB`
@PAR@ node-prop media.filename
File name for the stream
@PAR@ node-prop media.icon
Icon for the media, a base64 blob with PNG image data
@PAR@ node-prop media.icon-name
An XDG icon name for the media. Ex: `audio-x-mp3`
@PAR@ node-prop media.comment
Extra stream comment
@PAR@ node-prop media.date
Date of the media
@PAR@ node-prop media.format
User readable stream format information
@PAR@ node-prop object.linger = false
If the object should outlive its creator.
@PAR@ node-prop device.id
ID of the device the node belongs to.
## Classifying Properties @IDX@ props
The classifying properties of a node are use for routing the signal to its destination and
for configuring the settings.
@PAR@ node-prop media.type
The media type contains a broad category of the media that is being processed by the node.
Possible values include "Audio", "Video", "Midi"
@PAR@ node-prop media.category
\parblock
What kind of processing is done with the media. Possible values include:
* Playback: media playback.
* Capture: media capture.
* Duplex: media capture and playback or media processing in general.
* Monitor: a media monitor application. Does not actively change media data but monitors
activity.
* Manager: Will manage the media graph.
\endparblock
@PAR@ node-prop media.role
\parblock
The Use case of the media. Possible values include:
* Movie: Movie playback with audio and video.
* Music: Music listening.
* Camera: Recording video from a camera.
* Screen: Recording or sharing the desktop screen.
* Communication: VOIP or other video chat application.
* Game: Game.
* Notification: System notification sounds.
* DSP: Audio or Video filters and effect processing.
* Production: Professional audio processing and production.
* Accessibility: Audio and Visual aid for accessibility.
* Test: Test program.
\endparblock
@PAR@ node-prop media.class
\parblock
The media class is to classify the stream function. Possible values include:
* Video/Source: a producer of video, like a webcam.
* Video/Sink: a consumer of video, like a display window.
* Audio/Source: a source of audio samples like a microphone.
* Audio/Sink: a sink for audio samples, like an audio card.
* Audio/Duplex: a node that is both a sink and a source.
* Stream/Output/Audio: a playback stream.
* Stream/Input/Audio: a capture stream.
The session manager assigns special meaning to the nodes based on the media.class. Sink or Source
classes are used as targets for Stream classes, etc..
\endparblock
## Scheduling Properties @IDX@ props
@PAR@ node-prop node.latency = 1024/48000
Sets a suggested latency on the node as a fraction. This is just a suggestion, the graph will try to configure this latency or less for the graph. It is however possible that the graph is forced to a higher latency.
@PAR@ node-prop node.lock-quantum = false
\parblock
When this node is active, the quantum of the graph is locked and not allowed to change automatically.
It can still be changed forcibly with metadata or when a node forces a quantum.
JACK clients use this property to avoid unexpected quantum changes.
\endparblock
@PAR@ node-prop node.force-quantum = INTEGER
\parblock
While the node is active, force a quantum in the graph. The last node to be activated with this property wins.
A value of 0 unforces the quantum.
\endparblock
@PAR@ node-prop node.rate = RATE
Suggest a rate (samplerate) for the graph. The suggested rate will only be applied when doing so would not cause
interruptions (devices are idle) and when the rate is in the list of allowed rates in the server.
@PAR@ node-prop node.lock-rate = false
When the node is active, the rate of the graph will not change automatically. It is still possible to force a rate change with metadata or with a node property.
@PAR@ node-prop node.force-rate = RATE
\parblock
When the node is active, force a specific sample rate on the graph. The last node to activate with this property wins.
A RATE of 0 means to force the rate in `node.rate` denominator.
\endparblock
@PAR@ node-prop node.always-process = false
\parblock
When the node is active, it will always be joined with a driver node, even when nothing is linked to the node.
Setting this property to true also implies node.want-driver = true.
This is the default for JACK nodes, that always need their process callback called.
\endparblock
@PAR@ node-prop node.want-driver = true
The node wants to be linked to a driver so that it can start processing. This is the default for streams
and filters since 0.3.51. Nodes that are not linked to anything will still be set to the idle state,
unless node.always-process is set to true.
@PAR@ node-prop node.pause-on-idle = false
@PAR@ node-prop node.suspend-on-idle = false
\parblock
When the node is not linked anymore, it becomes idle. Normally idle nodes keep processing and are suspended by the session manager after some timeout. It is possible to immediately pause a node when idle with this property.
When the session manager does not suspend nodes (or when there is no session manager), the node.suspend-on-idle property can be used instead.
\endparblock
@PAR@ node-prop node.loop.name = null
@PAR@ node-prop node.loop.class = data.rt
\parblock
Add the node to a specific loop name or loop class. By default the node is added to the
data.rt loop class. You can make more specific data loops and then assign the nodes to those.
Other well known names are main-loop.0 and the main node.loop.class which runs the node data processing
in the main loop.
\endparblock
@PAR@ node-prop priority.driver # integer
\parblock
The priority of choosing this device as the driver in the graph. The driver is selected from all linked devices by selecting the device with the highest priority.
Normally, the session manager assigns higher priority to sources so that they become the driver in the graph. The reason for this is that adaptive resampling should be done on the sinks rather than the source to avoid signal distortion when capturing audio.
\endparblock
@PAR@ node-prop clock.name # string
\parblock
The name of the clock. This name is auto generated from the card index and stream direction. Devices with the same clock name will not use a resampler to align the clocks. This can be used to link devices together with a shared word clock.
In Pro Audio mode, nodes from the same device are assumed to have the same clock and no resampling will happen when linked together. So, linking a capture port to a playback port will not use any adaptive resampling in Pro Audio mode.
In Non Pro Audio profile, no such assumption is made and adaptive resampling is done in all cases by default. This can also be disabled by setting the same clock.name on the nodes.
\endparblock
## Session Manager Properties @IDX@ props
@PAR@ node-prop node.autoconnect = true
Instructs the session manager to automatically connect this node to some other node, usually
a sink or source.
@PAR@ node-prop node.exclusive = false
If this node wants to be linked exclusively to the sink/source.
This is a passive node and so it should not keep sinks/sources busy. This property makes the session manager create passive links to the sink/sources. If the node is not otherwise linked (via a non-passive link), the node and the sink it is linked to are idle (and eventually suspended).
This is used for filter nodes that sit in front of sinks/sources and need to suspend together with the sink/source.
\endparblock
@PAR@ node-prop node.link-group = ID
Add the node to a certain link group. Nodes from the same link group are not automatically linked to each other by the session manager. And example is a coupled stream where you don't want the output to link to the input streams, making a useless loop.
@PAR@ node-prop stream.dont-remix = false
Instruct the session manager to not remix the channels of a stream. Normally the stream channel configuration is changed to match the sink/source it is connected to. With this property set to true, the stream will keep its original channel layout and the session manager will link matching channels with the sink.
@PAR@ node-prop priority.session # integer
The priority for selecting this node as the default source or sink.
The merger is used as the input for a sink device node or a capture stream. It takes the various channels and merges them into a single stream for further processing.
The merger will also provide the monitor ports of the input channels and can
apply a software volume on the monitor signal.
@PAR@ node-prop monitor.channel-volumes = false
The volume of the input channels is applied to the volume of the monitor ports. Normally
the monitor ports expose the raw unmodified signal on the input ports.
## Resampler Parameters
Source, sinks, capture and playback streams contain a high quality adaptive resampler.
It uses [sinc](https://ccrma.stanford.edu/~jos/resample/resample.pdf) based resampling
with linear interpolation of filter banks to perform arbitrary
resample factors. The resampler is activated in the following cases:
* The hardware of a device node does not support the graph samplerate. Resampling will occur
from the graph samplerate to the hardware samplerate.
* The hardware clock of a device does not run at the same speed as the graph clock and adaptive
resampling is required to match the clocks.
* A stream does not have the same samplerate as the graph and needs to be resampled.
* An application wants to activate adaptive resampling in a stream to make it match some other
clock.
PipeWire performs most of the sample conversions and resampling in the client (Or in the case of the PulseAudio server, in the pipewire-pulse server that creates the streams). This ensures all the conversions are offloaded to the clients and the server can deal with one single format for performance reasons.
Below is an explanation of the options that can be tuned in the sample converter.
@PAR@ node-prop resample.quality = 4
\parblock
The quality of the resampler. from 0 to 14, the default is 4.
Increasing the quality will result in better cutoff and less aliasing at the expense of
(much) more CPU consumption. The default quality of 4 has been selected as a good compromise
between quality and performance with no artifacts that are well below the audible range.
See [Infinite Wave](https://src.infinitewave.ca/) for a comparison of the performance.
\endparblock
@PAR@ node-prop resample.disable = false
Disable the resampler entirely. The node will only be able to negotiate with the graph
when the samplerates are compatible.
## Channel Mixer Parameters
Source, sinks, capture and playback streams can apply channel mixing on the incoming signal.
Normally the channel mixer is not used for devices, the device channels are usually exposed as they are. This policy is usually enforced by the session manager, so we refer to its documentation there.
Playback and capture streams are usually configured to the channel layout of the sink/source
they connect to and will thus perform channel mixing.
The channel mixer also implements a software volume. This volume adjustment is performed on the original
channel layout. ex: A stereo playback stream that is up-mixed to 5.1 has 2 a left an right volume control.
@PAR@ node-prop channelmix.disable = false
Disables the channel mixer completely. The stream will only be able to link to compatible
sources/sinks with the exact same channel layout.
@PAR@ node-prop channelmix.min-volume = 0.0
@PAR@ node-prop channelmix.max-volume = 10.0
Gives the min and max volume values allowed. Any volume that is set will be clamped to these
values.
@PAR@ node-prop channelmix.normalize = false
\parblock
Makes sure that during such mixing & resampling original 0 dB level is preserved, so nothing sounds wildly quieter/louder.
While this options prevents clipping, it can in some cases produce too low volume. Increase the
volume in that case or disable normalization.
\endparblock
@PAR@ node-prop channelmix.lock-volumes = false
Completely disable volume or mute changes. Defaults to false.
@PAR@ node-prop channelmix.mix-lfe = true
Mixes the low frequency effect channel into the front center or stereo pair. This might enhance the dynamic range of the signal if there is no subwoofer and the speakers can reproduce the low frequency signal.
@PAR@ node-prop channelmix.upmix = true
\parblock
Enables up-mixing of the front center (FC) when the target has a FC channel.
The sum of the stereo channels is used and an optional lowpass filter can be used
(see `channelmix.fc-cutoff`).
Also enabled up-mixing of LFE when `channelmix.lfe-cutoff` is set to something else than 0 and
the target has an LFE channel. The LFE channel is produced by adding the stereo channels.
If `channelmix.upmix` is true, the up-mixing of the rear channels is also enabled and controlled
with the `channelmix-upmix-method` property.
\endparblock
@PAR@ node-prop channelmix.upmix-method = psd
\parblock
3 methods are provided to produce the rear channels in a surround sound:
1. none. No rear channels are produced.
2. simple. Front channels are copied to the rear. This is fast but can produce phasing effects.
3. psd. The rear channels as produced from the front left and right ambient sound (the
difference between the channels). A delay and optional phase shift are added to the rear signal
Apply a lowpass filter to the low frequency effects. The value is expressed in Hz. Typical subwoofers have a cutoff at around 150 and 200. The default value of 0 disables the feature.
@PAR@ node-prop channelmix.fc-cutoff = 12000
\parblock
Apply a lowpass filter to the front center frequency. The value is expressed in Hz.
Since the front center contains the dialogs, a typical cutoff frequency is 12000 Hz.
This option is only active when the up-mix is enabled.
\endparblock
@PAR@ node-prop channelmix.rear-delay = 12.0
\parblock
Apply a delay in milliseconds when up-mixing the rear channels. This improves
specialization of the sound. A typical delay of 12 milliseconds is the default.
This is only active when the `psd` up-mix method is used.
\endparblock
@PAR@ node-prop channelmix.stereo-widen = 0.0
\parblock
Subtracts some of the front center signal from the stereo channels. This moves the dialogs
more to the center speaker and leaves the ambient sound in the stereo channels.
This is only active when up-mix is enabled and a Front Center channel is mixed.
\endparblock
@PAR@ node-prop channelmix.hilbert-taps = 0
\parblock
This option will apply a 90 degree phase shift to the rear channels to improve specialization.
Taps needs to be between 15 and 255 with more accurate results (and more CPU consumption)
for higher values.
This is only active when the `psd` up-mix method is used.
\endparblock
@PAR@ node-prop dither.noise = 0
\parblock
This option will add N bits of random data to the signal. When no dither.method is
specified, the random data will flip between [-(1<<(N-1)), 0] every 1024 samples. With
a dither.method, the dither noise is amplified with 1<<(N-1) bits.
This can be used to keep some amplifiers alive during silent periods. One or two bits of noise is
usually enough, otherwise the noise will become audible. This is usually used together with
`session.suspend-timeout-seconds` to disable suspend in the session manager.
Note that PipeWire uses floating point operations with 24 bits precission for all of the audio
processing. Conversion to 24 bits integer sample formats is lossless and conversion to 32 bits
integer sample formats are simply padded with 0 bits at the end. This means that the dither noise
is always only in the 24 most significant bits.
\endparblock
@PAR@ node-prop dither.method = none
\parblock
Optional [dithering](https://en.wikipedia.org/wiki/Dither) can be done on the quantized
output signal.
There are 6 modes available:
1. none No dithering is done.
2. rectangular Dithering with a rectangular noise distribution. This adds random
bits in the [-0.5, 0.5] range to the signal with even distribution.
3. triangular Dithering with a triangular noise distribution. This add random
bits in the [-1.0, 1.0] range to the signal with triangular distribution
around 0.0.
4. triangular-hf Dithering with a sloped triangular noise distribution.
5. wannamaker3 Additional noise shaping is performed on the sloped triangular
dithering to move the noise to the more inaudible range. This is using
the "F-Weighted" noise filter described by Wannamaker.
6. shaped5 Additional noise shaping is performed on the triangular dithering
to move the noise to the more inaudible range. This is using the
Lipshitz filter.
Dithering is only useful for conversion to a format with less than 24 bits and will be
disabled otherwise.
\endparblock
## Debug Parameters
@PAR@ node-prop debug.wav-path = ""
Make the stream to also write the raw samples to a WAV file for debugging purposes.
## Other Parameters
These control low-level technical features:
@PAR@ node-prop clock.quantum-limit
\ref pipewire_conf__default_clock_quantum-limit "See pipewire.conf(5)"
@PAR@ node-prop resample.peaks = false # boolean
Instead of actually resampling, produce peak amplitude values as output.
This is used for volume monitoring, where it is set as a property
The number of audio channels to open the device with. Defaults depends on the profile of the device.
@PAR@ node-prop audio.rate # integer
The audio rate to open the device with. Default is 0, which means to open the device with a rate as close to the graph rate as possible.
@PAR@ node-prop audio.format # string
The audio format to open the device in. By default this is "UNKNOWN", which will open the device in the best possible bits (32/24/16/8..). You can force a format like S16_LE or S32_LE.
@PAR@ node-prop audio.position # JSON array of strings
The audio position of the channels in the device. This is auto detected based on the profile. You can configure an array of channel positions, like "[ FL, FR ]".
@PAR@ node-prop audio.allowed-rates # JSON array of integers
\parblock
The allowed audio rates to open the device with. Default is "[ ]", which means the device can be opened in any supported rate.
Only rates from the array will be used to open the device. When the graph is running with a rate not listed in the allowed-rates, the resampler will be used to resample to the nearest allowed rate.
\endparblock
@PAR@ node-prop api.alsa.period-size # integer
The period size to open the device in. By default this is 0, which will open the device in the default period size to minimize latency.
@PAR@ node-prop api.alsa.period-num # integer
The amount of periods to use in the device. By default this is 0, which means to use as many as possible.
@PAR@ node-prop api.alsa.headroom # integer
The amount of extra space to keep in the ringbuffer. The default is 0. Higher values can be configured when the device read and write pointers are not accurately reported.
Disable mmap operation of the device and use the ALSA read/write API instead. Default is false, mmap is preferred.
@PAR@ node-prop api.alsa.disable-batch # boolean
Ignore the ALSA batch flag. If the batch flag is set, ALSA will need an extra period to update the read/write pointers. Ignore this flag from ALSA can reduce the latency. Default is false.
@PAR@ node-prop api.alsa.use-chmap # boolean
Use the driver provided channel map. Default is true when using UCM, false otherwise because many driver don't report this correctly.
@PAR@ node-prop api.alsa.multi-rate # boolean
Allow devices from the same card to be opened in multiple sample rates. Default is true. Some older drivers did not properly advertise the capabilities of the device and only really supported opening the device in one rate.
These settings are not saved and need to be reapplied for each session manager restart.
# ALSA CARD PROFILES @IDX@ props
The sound card profiles ("Analog stereo", "Analog stereo duplex", ...) except "Pro Audio" come from two sources:
- UCM: ALSA Use Case Manager: the profile configuration system from ALSA. See https://github.com/alsa-project/alsa-ucm-conf/
- ACP ("Alsa Card Profiles"): Pulseaudio's profile system ported to PipeWire. See https://www.freedesktop.org/wiki/Software/PulseAudio/Backends/ALSA/Profiles/
See the above links on how to configure these systems.
For ACP, PipeWire looks for the profile configuration files under
