shairport-sync.7.xml

<?xml version="1.0"?><!--*-nxml-*-->
<!DOCTYPE manpage SYSTEM "xmltoman.dtd">
<?xml-stylesheet type="text/xsl" href="xmltoman.xsl" ?>

  <!--
  This man page source file is part of shairport-sync.
  Copyright (c) Mike Brady 2014-2018

  All rights reserved.
 
  Permission is hereby granted, free of charge, to any person
  obtaining a copy of this software and associated documentation
  files (the "Software"), to deal in the Software without
  restriction, including without limitation the rights to use,
  copy, modify, merge, publish, distribute, sublicense, and/or
  sell copies of the Software, and to permit persons to whom the
  Software is furnished to do so, subject to the following conditions:
 
  The above copyright notice and this permission notice shall be
  included in all copies or substantial portions of the Software.
 
  THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
  EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES
  OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
  NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
  HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
  WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
  FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  OTHER DEALINGS IN THE SOFTWARE.
  -->

  <manpage name="shairport-sync" section="7" desc="Synchronised Audio Player for iTunes / AirPlay">

	<synopsis>
	<!--
      <cmd>shairport-sync <opt>-D</opt> <arg>interface</arg></cmd>
  -->
      <cmd>shairport-sync  <opt>[-djvw]</opt>
      <opt>[-a </opt><arg>name</arg><opt>]</opt>
      <opt>[-A </opt><arg>latency</arg><opt>]</opt>
      <opt>[-B </opt><arg>command</arg><opt>]</opt>
      <opt>[-c </opt><arg>configurationfile</arg><opt>]</opt>
      <opt>[-E </opt><arg>command</arg><opt>]</opt>
      <opt>[--get-cover-art]</opt>
      <opt>[--logOutputLevel]</opt>
      <opt>[-L </opt><arg>latency</arg><opt>]</opt>
      <opt>[-m </opt><arg>backend</arg><opt>]</opt>
      <opt>[--meta-dir=</opt><arg>directory</arg><opt>]</opt>
      <opt>[-o </opt><arg>backend</arg><opt>]</opt>
      <opt>[--password=</opt><arg>secret</arg><opt>]</opt>
      <opt>[-r </opt><arg>threshold</arg><opt>]</opt>
      <opt>[--statistics]</opt>
      <opt>[-S </opt><arg>mode</arg><opt>]</opt>
      <opt>[-t </opt><arg>timeout</arg><opt>]</opt>
      <opt>[--tolerance=</opt><arg>frames</arg><opt>]</opt>
      <opt>[-- </opt><arg>audio_backend_options</arg><opt>]</opt>
      </cmd>
      <cmd>shairport-sync <opt>-D</opt></cmd>
      <cmd>shairport-sync <opt>-k</opt></cmd>
      <cmd>shairport-sync <opt>-h</opt></cmd>
      <cmd>shairport-sync <opt>-R</opt></cmd>
      <cmd>shairport-sync <opt>-V</opt></cmd>
	</synopsis>

  <description>
    <p>shairport-sync  plays  audio  streamed  from  iTunes or from an AirPlay
    device to an ALSA compatible audio output device (available on Linux and FreeBSD) , to a "sndio" output device (available on OpenBSD, FreeBSD and Linux) or to a PulseAudio output stream (available on Linux).</p>
    <p>A feature of shairport-sync is that it offers full audio synchronisation.
    Full audio synchronisation means that audio is played on the output device at exactly the time specified by the audio source.
    This means that if many devices are playing the same stream at the same
    time, all the outputs will stay in step with one another.
    This allows multiple devices to play the same source without getting out of phase with one another,
    enabling, for example, simultaneous multi-room operation.
    </p>
    
    <p>shairport-sync can be compiled to stream audio synchronised audio output to a unix pipe or to standard output, or to audio systems that do not provide timing information. This could perhaps be described as partial audio synchronisation, where synchronised audio is provided by shairport-sync, but what happens to it in the subsequent processing chain, before it reaches the listener's ear, is outside the control of shairport-sync.</p>
    <p>shairport-sync can be compiled to stream metadata to a pipe or socket.</p>
    
    <p>Settings can be made using the configuration file (recommended for all new installations) or by using command-line options.</p>
 
	</description>
	
	<section name="Configuration File Settings">
	   <p>You should use the configuration file for setting up shairport-sync.
	  This file is usually <file>shairport-sync.conf</file> and is generally located in the System Configuration Directory, which is normally the <file>/etc</file> directory in Linux or the <file>/usr/local/etc</file> directory in BSD unixes.
	  You may need to have root privileges to modify it.</p>
	  
	  <p>(Note: Shairport Sync may have been compiled to use a different configuration directory. You can determine which by performing the command <file>$ shairport-sync -V</file>. One of the items in the output string is the value of the <opt>sysconfdir</opt>,
	  i.e. the System Configuration Directory.)</p>
	  
	  
	  <p>Within the configuraton file, settings are organised into <i>groups</i>, for example, there is a "general" group of
	  standard settings, and there is an "alsa" group with settings that pertain to the ALSA
	  back end. Here is an example of a typical configuration file:</p>
	  
	  <p><opt>general = {</opt></p>
    <p><p><opt>name = "Mike's Boombox";</opt></p></p>
    <p><p><opt>interpolation = "soxr";</opt></p></p>
    <p><p><opt>password = "secret";</opt></p></p>
    <p><p><opt>output_backend = "alsa";</opt></p></p>
    <p><opt>};</opt></p>
    <p><opt></opt></p>
    <p><opt>alsa = {</opt></p>
    <p><p><opt>output_device = "hw:0";</opt></p></p>
    <p><p><opt>mixer_control_name = "PCM";</opt></p></p>
    <p><opt>};</opt></p>
	  
	  <p>Most settings have sensible default values, so -- as in the example above -- users generally only need to set (1) the service name, (2) a password (if desired) and
	  (3) the output device. If the output device has a mixer that can be used for volume control, then (4) the volume control's name should be specified. It is highly desirable to use the output device's mixer for volume control, if available -- response time is reduced to zero and the processor load is reduced. In the example above, "soxr" interpolation was also enabled.</p>
	  
	  <p>A sample configuration file with all possible settings, but with all of them commented out, is installed at <file>shairport-sync.conf.sample</file>, within the System Configuration Directory -- <file>/etc</file> in Linux, <file>/usr/local/etc</file> in BSD unixes.</p>
	  
	  <p>To retain backwards compatibility with previous versions of shairport-sync
	  you can use still use command line options, but any new features, etc. will
	  be available only via configuration file settings.</p>

	  <p>The configuration file is processed using the <file>libconfig</file> library
	  -- see <url href="http://www.hyperrealm.com/libconfig/libconfig_manual.html"/>.</p>

    <option><p><opt>"GENERAL" SETTINGS</opt></p></option>
    <p>These are the settings available within the <opt>general</opt> group:</p>

    <option>
		<p><opt>name=</opt><arg>"service_name"</arg><opt>;</opt></p>
		<optdesc>
		<p>Use this <arg>service_name</arg> to identify this player in iTunes, etc.</p>
		<p>The following substitutions are allowed:
		<opt>%h</opt> for the computer's hostname,
		<opt>%H</opt> for the computer's hostname with the first letter capitalised (ASCII only),
		<opt>%v</opt> for the shairport-sync version number, e.g. "3.0.1" and
		<opt>%V</opt> for the shairport-sync version string, e.g. "3.0.1-OpenSSL-Avahi-ALSA-soxr-metadata-sysconfdir:/etc".</p>
    <p>The default is "%H", which is replaced by the hostname with the first letter capitalised.</p>
    </optdesc>
	  </option>
	  
	  <option>
    <p><opt>password=</opt><arg>"password"</arg><opt>;</opt></p>
    <optdesc><p>Require the password <arg>password</arg> to connect to the service. If you leave this setting commented out, no password is needed.</p></optdesc>
    </option>
    <option>
    <p><opt>interpolation=</opt><arg>"mode"</arg><opt>;</opt></p>
    <optdesc><p>Interpolate, or "stuff", the audio stream using the <arg>mode</arg>.  Interpolation here refers to the
    process of adding or removing frames of audio  to  or  from  the
    stream sent to the output device to keep it exactly in synchrony
    with the player.
    The default mode, "basic", is normally almost  completely  inaudible.
    The  alternative mode, "soxr", is even less obtrusive but
    requires much more processing power. For this mode, support for
    libsoxr, the SoX Resampler Library, must be selected when
    shairport-sync is compiled.
		</p></optdesc>
    </option>
    
    <option>
    <p><opt>statistics=</opt><arg>"setting"</arg><opt>;</opt></p>
    <optdesc><p>Use this <arg>setting</arg> to enable ("yes") or disable ("no") the output of some statistical information on the console or in the log. The default is to disable statistics.</p></optdesc>
    </option>

    <option>
    <p><opt>mdns_backend=</opt><arg>"backend"</arg><opt>;</opt></p>
    <optdesc><p>shairport-sync has a number of modules of code ("backends") for interacting with the mDNS service to be used to advertise itself. Normally, the first mDNS backend that works is selected. This setting forces the selection of the specific mDNS <arg>backend</arg>. The default is "avahi". Perform the command <opt>shairport-sync -h</opt> to get a list of available mDNS modules.</p></optdesc>
    </option>
    <option>
    <p><opt>output_backend=</opt><arg>"backend"</arg><opt>;</opt></p>
    <optdesc><p>shairport-sync has a number of modules of code ("backends") through which audio is output. Normally, the first audio backend that works is selected. This setting forces the selection of the specific audio <arg>backend</arg>. Perform the command <opt>shairport-sync -h</opt> to get a list of available audio backends -- the default is the first on this list. Only the "alsa", "sndio" and "pa" backends support synchronisation.</p></optdesc>
    </option>
    <option>
    <p><opt>port=</opt><arg>portnumber</arg><opt>;</opt></p>
    <optdesc><p>Use this to specify the <arg>portnumber</arg> shairport-sync uses to listen for service requests from iTunes, etc. The default is port 5000.</p></optdesc>
    </option>
    <option>
    <p><opt>udp_port_base=</opt><arg>portnumber</arg><opt>;</opt></p>
    <optdesc><p>When shairport-sync starts to play audio, it establises three UDP connections to the audio source. Use this setting to specify the starting <arg>portnumber</arg> for these three ports. It will pick the first three unused ports starting from <arg>portnumber</arg>. The default is port 6001.</p></optdesc>
    </option>
    <option>
    <p><opt>udp_port_range=</opt><arg>range</arg><opt>;</opt></p>
    <optdesc><p>Use this in conjunction with the prevous setting to specify the <arg>range</arg> of ports that can be checked for availability. Only three ports are needed.
    The default is 100, thus 100 ports will be checked from port 6001 upwards until three are found.</p></optdesc>
    </option>
    <option>
    <p><opt>drift_tolerance_in_seconds=</opt><arg>seconds</arg><opt>;</opt></p>
    <optdesc><p>Allow playback to drift up to <arg>seconds</arg> out of exact synchronization before attempting to correct it.
		The default is 0.002 seconds, i.e. 2 milliseconds. The smaller the tolerance, the more likely it is that overcorrection will occur.
		Overcorrection is when more corrections (insertions and deletions) are made than are strictly necessary to keep the stream in sync. Use the <opt>statistics</opt> setting to
		monitor correction levels. Corrections should not greatly exceed net corrections. This setting replaces the deprecated <opt>drift</opt> setting.
		</p></optdesc>
    </option>
    <option>
    <p><opt>resync_threshold_in_seconds=</opt><arg>threshold</arg><opt>;</opt></p>
    <optdesc><p>Resynchronise if timings differ by more than <arg>threshold</arg> seconds.
    If the output timing differs from the source timing by more than
    the threshold, output will be muted and a full resynchronisation
    will occur. The default threshold is 0.050 seconds, i.e. 50
    milliseconds. Specify 0.0 to disable resynchronisation.
    This setting replaces the deprecated <opt>resync_threshold</opt> setting.
    </p></optdesc>
    </option>
    <option>
    <p><opt>log_verbosity=</opt><arg>0</arg><opt>;</opt></p>
    <optdesc><p>Use this to specify how much debugging information should be output or logged. The value <arg>0</arg> means no debug information, <arg>3</arg> means most debug information. The default is <arg>0</arg>.</p></optdesc>
    </option>
    <option>
    <p><opt>ignore_volume_control=</opt><arg>"choice"</arg><opt>;</opt></p>
    <optdesc><p>Set this <arg>choice</arg> to <arg>"yes"</arg> if you want the volume to be at 100% no matter what the source's volume control is set to.
    This might be useful if you want to set the volume on the output device, independently of the setting at the source. The default is <arg>"no"</arg>.</p></optdesc>
    </option>
    
    <option>
    <p><opt>volume_max_db=</opt><arg>dBvalue</arg><opt>;</opt></p>
    <optdesc><p>Specify the maximum output level to be used with the hardware mixer, if used. If no hardware mixed is used, this setting speciies the maximum setting permissible in the software mixer, which has an attenuation of from 0.0 dB down to -96.3 dB.
    </p></optdesc>
    </option>
    
    <option>
    <p><opt>volume_range_db=</opt><arg>dBvalue</arg><opt>;</opt></p>
    <optdesc><p>Use this <arg>dBvalue</arg> to reduce or increase the attenuation range, in decibels, between the minimum and maximum volume.</p>
    
    <p>For example, if a mixer has a minimum volume of -80 dB and a maximum of +20 dB, you might wish to use only 60 dB of the 100 dB available.
    This might be because the sound becomes inaudible at the lowest setting and unbearably loud at the highest setting --
    indeed, many domestic HiFi systems have a volume control range of just 60 to 80dB.</p>
    <p>Another potential use might be where the range specified by the mixer does not match the capabilities of the device.
    For example, the Raspberry Pi's DAC that feeds the built-in audio jack claims a range of 106 dB but has a useful range of only about 30 dB.
    The setting allows you to specify the maximum range from highest to lowest.
    The range suggested for the Raspberry Pi's built-in audio DAC, which feeds the headphone jack, is 30.
    Using it in this case gives the volume control a much more useful range of settings.</p>
    <p>As a third example, you can actually extend the range provided by a mixer.
    Many cheaper DACs have hardware mixers that offer a restricted attenuation range.
    If you specify a volume range greater than the range of the mixer, software attenuation and hardware attenuation
    will be combined to give the specified range.</p>
    <p>If you omit this setting, the native range of the mixer is used.</p></optdesc>
    </option>
    <option>
    <p><opt>regtype=</opt><arg>"regTypeString"</arg><opt>;</opt></p>
    <optdesc><p>Use this advanced setting to set the service type and transport to be advertised by Zeroconf/Bonjour. Default is <arg>"_raop._tcp"</arg>.</p></optdesc>
    </option>
    
    <option>
    <p><opt>playback_mode=</opt><arg>"mode"</arg><opt>;</opt></p>
    <optdesc><p>The <arg>mode</arg> can be "stereo", "mono", "reverse stereo", "both left" or "both right". Default is "stereo".</p></optdesc>
    </option>
    
    <option>
    <p><opt>interface=</opt><arg>"name"</arg><opt>;</opt></p>
    <optdesc><p>Use this advanced setting if you want to confine Shairport Sync to the named interface. Leave it commented out to get the default bahaviour.</p></optdesc>
    </option>
    
    <option>
    <p><opt>alac_decoder=</opt><arg>"decodername"</arg><opt>;</opt></p>
    <optdesc><p>This can be "hammerton" or "apple". This advanced setting allows you to choose the original Shairport decoder by David Hammerton or the Apple Lossless Audio Codec (ALAC) decoder written by Apple. Shairport Sync must have been compiled with the configuration setting "--with-apple-alac" and the Apple ALAC decoder library must be present for this to work.</p></optdesc>
    </option>
    <option>
    <p><opt>audio_backend_latency_offset_in_seconds=</opt><arg>offset_in_seconds</arg><opt>;</opt></p>
    <optdesc><p>Set this <arg>offset_in_seconds</arg> to compensate for a fixed delay in the audio back end.
    For example, if the output device delays by 100 ms, set this to -0.1.</p></optdesc>
    </option>
    <option>
    <p><opt>audio_backend_buffer_desired_length_in_seconds=</opt><arg>length_in_seconds</arg><opt>;</opt></p>
    <optdesc><p>Use this <arg>length_in_seconds</arg> to set the desired length of the queue of audio frames in the backend's output buffer.</p>
    <p>The default is 0.15 seconds for the ALSA backend, 0.35 seconds for the PA backend and one second for all other backends.</p>
    <p>If this value is set too small, underflow may occur on low-powered machines.
    If set too large, the response times to the volume control may become excessive, or it may exceed the backend's buffer size.
    It may need to be larger on low-powered machines that are also performing other tasks, such as processing metadata.</p></optdesc>
    </option>
    <option>
    <p><opt>audio_backend_silent_lead_in_time=</opt><arg>lead_in_time_in_seconds</arg><opt>;</opt></p>
    <optdesc><p>This is an advanced setting. Use the <arg>lead_in_time_in_seconds</arg> to set the desired length of the period of silence
    (a "silent lead-in") played before a play session begins.</p>
    <p>The purpose of this silent lead-in is to give the backend sufficient time to prepare for operation and to make an estimate
    (and, importantly,  to correct the estimate) of the exact time at which to begin playing audio to achieve initial synchronisation.
    The value can be from 0.0 up to a maximum of either 4.0 seconds. The actual duration will be close to the setting but can not exceed the latency set by the client,
    usually 2 seconds or a little more.</p>
    <p>If the value chosen is too short for synchronised backends such as the ALSA, sndio or PA backends, then audio will not be synchronised correctly at the start of play.
    The default is to have a silent lead-in of approximately the same time as the latency set by the client.</p></optdesc>
    </option>
    <option>
    <p><opt>run_this_when_volume_is_set=</opt><arg>"/full/path/to/application/and/args"</arg><opt>;</opt></p>
    <optdesc><p>Here you can specify a program and its arguments that will be run when the volume is set or changed. Be careful to include the full path to the application.
    The application must be marked as executable and, if it is a script, its first line must begin with the standard shebang <file>#!/bin/...</file> as appropriate.</p>
    <p>The desired AirPlay volume is appended to the end of the command line - leave a space if you want it treated as an extra argument.
    AirPlay volume goes from 0.0 to -30.0 and -144.0 means "mute".</p></optdesc>
    </option>


    <option><p><opt>"ALSA" SETTINGS</opt></p></option>
    <p>These settings are for the ALSA back end, used to communicate with audio output devices in the ALSA system.
    (By the way, you can use tools such as <opt>alsamixer</opt> or <opt>aplay</opt> to discover what devices are available.)
    Use these settings to select the output device and the mixer control to be used to control the output volume.
    You can additionally set the desired size of the output buffer and you can adjust overall latency. Here are the <opt>alsa</opt> group settings:</p>

    <option>
    <p><opt>output_device=</opt><arg>"output_device"</arg><opt>;</opt></p>
    <optdesc><p>Use the output device called <arg>output_device</arg>. The default is the device called <arg>"default"</arg>.</p></optdesc>
    </option>
    <option>
    <p><opt>mixer_control_name=</opt><arg>"name"</arg><opt>;</opt></p>
    <optdesc><p>Specify the <arg>name</arg> of the mixer control to be used by shairport-sync to control the volume.
    The mixer control must be on the mixer device, which by default is the output device.
    If you do not specify a mixer control name, shairport-sync will adjust the volume in software.</p></optdesc>
    </option>
    <option>
    <p><opt>mixer_device=</opt><arg>"mixer_device"</arg><opt>;</opt></p>
    <optdesc><p>By default, the mixer is assumed to be output_device. Use this setting to specify a device other than the output device.</p></optdesc>
    </option>
    
    <option>
    <p><opt>output_rate=</opt><arg>frame rate</arg><opt>;</opt></p>
    <optdesc><p>Use this setting to specify the frame rate to output to the ALSA device. Allowable values are 44100 (default), 88200, 176400 and 352800. The device must have the capability to accept the format you specify. There is no particular reason to use anything other than 44100 if it is available.
    </p></optdesc>
    </option>

    <option>
    <p><opt>output_format=</opt><arg>"format"</arg><opt>;</opt></p>
    <optdesc><p>Use this setting to specify the format that should be used to send data to the ALSA device. Allowable values are "U8", "S8", "S16", "S24", "S24_3LE", "S24_3BE" or "S32". The device must have the capability to accept the format you specify.</p><p>"S" means signed; "U" means unsigned; BE means big-endian and LE means little-endian. Except where stated (using *LE or *BE), endianness matches that of the processor. The default is "S16".</p><p>If you are using a hardware mixer, the best setting is S16, as audio will pass through Shairport Sync unmodifed except for interpolation. If you are using the software mixer, use 32- or 24-bit, if your device is capable of it, to get the lowest possible levels of dither.
    </p></optdesc>
    </option>

    <option>
    <p><opt>disable_synchronization=</opt><arg>"no"</arg><opt>;</opt></p>
    <optdesc><p>This is an advanced setting and is for debugging only. Set to <arg>"yes"</arg> to disable synchronization. Default is <arg>"no"</arg>.
    If you use it to disable synchronisation, then sooner or later you'll experience audio glitches due to
    audio buffer overflow or underflow.
    </p></optdesc>
    </option>
    <option>
    <p><opt>period_size=</opt><arg>number</arg><opt>;</opt></p>
    <optdesc><p>Use this optional advanced setting to set the alsa period size near to this value.</p></optdesc>
    </option>
    <option>
    <p><opt>buffer_size=</opt><arg>number</arg><opt>;</opt></p>
    <optdesc><p>Use this optional advanced setting to set the alsa buffer size near to this value.</p></optdesc>
    </option>
    <option>
    <p><opt>use_mmap_if_available=</opt><arg>"yes"</arg><opt>;</opt></p>
    <optdesc><p> Use this optional advanced setting to control whether MMAP-based output is used to communicate with the DAC. Default is <arg>"yes"</arg>.</p></optdesc>
    </option>
    
    <option>
    <p><opt>mute_using_playback_switch=</opt><arg>"no"</arg><opt>;</opt></p>
    <optdesc>
    <p>This is an advanced setting and the default is <arg>"no"</arg>. If it is set to <arg>"yes"</arg>, hardware mute will be implemented using a feature called a 'playback switch', where one is available. Set it to <arg>"no"</arg> to prevent the playback switch being used.</p>
    <p>If Shairport Sync is sharing the output device with other applications, it is best to leave this set to <arg>"no"</arg> for compatibility with those applications.</p>
    <p>Another motivation for this is to allow the alsa function call
    "snd_mixer_selem_set_playback_switch_all" to be avoided. It is incorrectly implemented on certain soundcards, including the emulated card in VMWare Fusion 8.5.</p>
    </optdesc>
    </option>
    <option><p><opt>"SNDIO" SETTINGS</opt></p></option>
     <p>These settings are for the SNDIO back end, used to communicate with audio output devices in the SNDIO system.</p>
    <option>
    <p><opt>device=</opt><arg>"snd/0"</arg><opt>;</opt></p>
    <optdesc><p>Use this optional setting to specify the name of the output device, e.g. <arg>"snd/0"</arg>. The default is to use the SNDIO system's default.</p></optdesc>
    </option>
    <option>
    <p><opt>rate=</opt><arg>44100</arg><opt>;</opt></p>
    <optdesc><p>Use this optional setting to specify the output rate in frames per second. Valid rates are 44100, 88200, 176400 or 352800.
    The output device must have the capability to accept data at the specified rate. The default is 44100.</p></optdesc>
    </option>
    <option>
    <p><opt>format=</opt><arg>"S16"</arg><opt>;</opt></p>
    <optdesc><p>Use this optional setting to specify the output format.  Allowable values are "U8", "S8", "S16", "S24", "S24_3LE", "S24_3BE" or "S32".
    The device must have the capability to accept the format you specify.</p><p>"S" means signed; "U" means unsigned; BE means big-endian and LE means little-endian.
    Except where stated (using *LE or *BE), endianness matches that of the processor. The default is "S16".</p><p>
    Since the SNDIO backend does not use a hardware mixer for volume control, dither will be introduced into the output if it is less than full volume.
    Thus, (unless you are ignoring the volume control setting),
    consider using 32- or 24-bit output if your device is capable of it, to get the lowest possible levels of dither.</p>
    <p>Please note that 32- or 24-bit has not been extensively tested on SNDIO.</p></optdesc>
    </option>
    <option>
    <p><opt>round=</opt><arg>value</arg><opt>;</opt></p>
    <optdesc><p>Use this optional advanced setting to specify the period size of the SNDIO channel. If omitted, a SNDIO system default value will be used.</p></optdesc>
    </option>
    <option>
    <p><opt>bufsiz=</opt><arg>value</arg><opt>;</opt></p>
    <optdesc><p>Use this optional advanced setting to specify the buffer size of the SNDIO channel. If omitted, a SNDIO system default value will be used.</p></optdesc>
    </option>


    <option><p><opt>"PA" SETTINGS</opt></p></option>
    <p>These settings are for the new PulseAudio backend.</p>
    <option>
    <p><opt>application_name=</opt><arg>"Shairport Sync"</arg><opt>;</opt></p>
    <optdesc><p>Use this to set the name to appear in the Sounds "Applications" tab when Shairport Sync is active. The default is the name "Shairport Sync".</p></optdesc>
    </option>

    <option><p><opt>"PIPE" SETTINGS</opt></p></option>
    <p>These settings are for the PIPE backend, used to route audio to a named unix pipe. The audio is in raw CD audio format: PCM 16 bit little endian, 44,100 samples per second,
    interleaved stereo.</p>
    <option>
    <p><opt>name=</opt><arg>"/path/to/pipe"</arg><opt>;</opt></p>
    <optdesc><p>Use this to specify the name and location of the pipe. The pipe will be created and opened when shairport-sync starts up
    and will be closed upon shutdown.
    Frames of audio will be sent to the pipe in packets of 352 frames and will be discarded if the pipe has not have a reader attached.
    The sender will wait for up to five seconds for a packet to be written before discarding it.</p></optdesc>
    </option>
     
    <option><p><opt>"STDOUT" SETTINGS</opt></p></option>
    <p>There are no settings for the STDOUT backend.</p>
    
    <option><p><opt>"AO" SETTINGS</opt></p></option>
    <p>There are no configuration file settings for the AO backend.</p>

    
    <option><p><opt>"METADATA" SETTINGS</opt></p></option>
    <p>shairport-sync can process metadata provided by the source, such as Track Number, Album Name, cover art, etc. and can provide additional metadata such as volume level,
    pause/resume, etc. It sends the metadata to a pipe, by default <file>/tmp/shairport-sync-metadata</file>.
    To process metadata, shairport-sync must have been compiled with metadata support included.
    You can check that this is so by running the command <opt>$ shairport-sync -V</opt>; the identification string will contain the word <opt>metadata</opt>.</p>
    <p>Please note that different sources provide different levels of metadata. Some provide a lot; some provide almost none.</p>
    <p>The <opt>metadata</opt> group of settings allow you to enable metadata handling and to control certain aspects of it:</p>


    <option>
    <p><opt>enabled=</opt><arg>"choice"</arg><opt>;</opt></p>
    <optdesc><p>Set the <arg>choice</arg> to "yes" to enable shairport-sync to look for metadata from the audio source and to forward it,
    along with metadata generated by shairport-sync itself, to the metadata pipe. The default is "no".</p></optdesc>
    </option>
    <option>
    <p><opt>include_cover_art=</opt><arg>"choice"</arg><opt>;</opt></p>
    <optdesc><p>Set the <arg>choice</arg> to "yes" to enable shairport-sync to look for cover art from the audio source and to include it in the feed to the metadata pipe.
    You must also enable metadata (see above).
    One reason for not including cover art is that the images can sometimes be very large and may delay transmission of subsequent metadata through the pipe.
    The default is "no".</p></optdesc>
    </option>
    <option>
    <p><opt>pipe_name=</opt><arg>"filepathname"</arg><opt>;</opt></p>
    <optdesc><p>Specify the absolute path name of the pipe through which metadata should be sent The default is <file>/tmp/shairport-sync-metadata</file>.</p></optdesc>
    </option>
    
    <option>
    <p><opt>socket_address=</opt><arg>"hostnameOrIP"</arg><opt>;</opt></p>
    <optdesc><p>If <arg>hostnameOrIP</arg> is set to a host name or and IP address, UDP packets containing metadata will be sent to this address.
    May be a multicast address. Additionally, <arg>socket-port</arg> must be non-zero and <arg>enabled</arg> must be set to "yes".</p></optdesc>
    </option>
    <option>
    <p><opt>socket_port=</opt><arg>port</arg><opt>;</opt></p>
    <optdesc><p>If <opt>socket_address</opt> is set, use <arg>port</arg> to specify the port to send UDP packets to. Must not be zero.</p></optdesc>
    </option>
    <option>
    <p><opt>socket_msglength=</opt><arg>65000</arg><opt>;</opt></p>
    <optdesc><p>The maximum packet size for any UDP metadata. This must be between 500 or 65000. The default is 500.</p></optdesc>
    </option>

    <option><p><opt>"SESSIONCONTROL" SETTINGS</opt></p></option>
    <p>shairport-sync can run programs just before it starts to play an audio stream and just after it finishes.
    You specify them using the sessioncontrol group settings run_this_before_play_begins and run_this_after_play_ends.</p>

    <option>
    <p><opt>run_this_before_play_begins=</opt><arg>"/path/to/application and args"</arg><opt>;</opt></p>
    <optdesc><p>Here you can specify a program and its arguments that will be run just before a play session begins. Be careful to include the full path to the application.
    The application must be marked as executable and, if it is a script, its first line must begin with the standard shebang <file>#!/bin/...</file> as appropriate.</p></optdesc>
    </option>
    <option>
    <p><opt>run_this_after_play_ends=</opt><arg>"/path/to/application and args"</arg><opt>;</opt></p>
    <optdesc><p>Here you can specify a program and its arguments that will be run just after a play session ends. Be careful to include the full path to the application.
    The application must be marked as executable and, if it is a script, its first line must begin with the standard shebang <file>#!/bin/...</file> as appropriate.</p></optdesc>
    </option>
    <option>
    <p><opt>wait_for_completion=</opt><arg>"choice"</arg><opt>;</opt></p>
    <optdesc><p>Set <arg>choice</arg> to "yes" to make shairport-sync wait until the programs specified in the <opt>run_this_before_play_begins</opt>,
    <opt>run_this_after_play_ends</opt> and <opt>run_this_when_volume_is_set</opt> have completed execution before continuing. The default is "no".</p></optdesc>
    </option>
    <option>
    <p><opt>allow_session_interruption=</opt><arg>"choice"</arg><opt>;</opt></p>
    <optdesc><p>If <opt>choice</opt> is set to "yes", then another source will be able to interrupt an existing play session and start a new one.
    When set to "no" (the default), other devices attempting to interrupt a session will fail, receiving a busy signal.</p></optdesc>
    </option>
    <option>
    <p><opt>session_timeout=</opt><arg>seconds</arg><opt>;</opt></p>
    <optdesc><p>If a play session has been established and the source disappears without warning (such as a device going out of range of a network)
    then wait for <arg>seconds</arg> seconds before ending the session. Once the session has terminated, other devices can use it.
    The default is 120 seconds.</p></optdesc>
    </option>
    
	</section>

	<options>
	  <p>This section is about the command-line options available in shairport-sync.</p>

    <p>Note: if you are setting up shairport-sync for the first time or are updating an existing installation,
    you are encouraged to use the configuration file settings described above. Most of the command-line options described below
    simply replicate the configuration settings and are retained to provide backward compatibility with older installations of shairport-sync.</p>
    
    <p>Many  command-line options take sensible default values, so you can normally
    ignore most of them. See the EXAMPLES section for typical usages.</p>

    <p>There are two kinds of command-line options for shairport-sync:
    regular <opt>program options</opt> and <opt>audio backend options</opt>.
    Program options are
    always listed first, followed by any audio backend options, preceded by
    a <opt>--</opt> symbol.</p>
    
  <section name="Program Options">
  <p>These command-line options are used by shairport-sync itself.</p>
  </section>
  
	  <option>
		<p><opt>-a </opt><arg>service name</arg><opt> | --name=</opt><arg>service name</arg></p>
		<optdesc><p>
		Use this <arg>service name</arg> to identify this player in iTunes, etc.</p>
		
		<p>The following substitutions are allowed:
		<opt>%h</opt> for the computer's hostname,
		<opt>%H</opt> for the computer's hostname with the first letter capitalised (ASCII only),
		<opt>%v</opt> for the shairport-sync version number, e.g. "3.0.1" and
		<opt>%V</opt> for the shairport-sync version string, e.g. "3.0.1-OpenSSL-Avahi-ALSA-soxr-metadata-sysconfdir:/etc".</p> 
    <p>The default is "%H", which is replaced by the hostname with the first letter capitalised.</p>
    </optdesc>
	  </option>

	  <option>
		<p><opt>-B </opt><arg>program</arg><opt> | --on-start=</opt><arg>program</arg></p>
		<optdesc><p>
		Execute <arg>program</arg> when playback is about to begin. Specify the 
		full path to the program, e.g. <file>/usr/bin/logger</file>.
		Executable scripts can be used, but they must have the appropriate shebang
    (<file>#!/bin/sh</file> in the headline.</p>
		
		<p>If you want shairport-sync to wait until the command has
		completed before starting to play, select the <opt>-w</opt> option as well.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-c </opt><arg>filename</arg><opt> | --configfile=</opt><arg>filename</arg></p>
		<optdesc><p>
		Read configuration settings from <arg>filename</arg>. The default is to read them from the <file>shairport-sync.conf</file> in the System Configuration Directory -- <file>/etc</file> in Linux, <file>/usr/local/etc</file> in BSD unixes.
		For information about configuration settings, see the  "Configuration File Settings" section above.
    </p></optdesc>
	  </option>

	  <option>
		<p><opt>-D | --disconnectFromOutput</opt></p>
		<optdesc><p>
		Disconnect the shairport-sync daemon from the output device and
    exit. (Requires that the daemon has written its PID to an agreed
    file -- see the <opt>-d</opt> option).
		</p>
		<p>Please note that this feature is deprecated and will be removed in a future version of shairport-sync.</p>
		</optdesc>
	  </option>

	  <option>
		<p><opt>-d | --daemon</opt></p>
		<optdesc><p>
		Instruct shairport-sync to demonise itself. It will write its
    Process ID (PID) to a file, usually at
    <file>/var/run/shairport-sync/shairport-sync.pid</file>, which is used by the
    <opt>-k</opt>, <opt>-D</opt> and <opt>-R</opt> options to locate
    the daemon at a later time. See also the <opt>-j</opt> option.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-E </opt><arg>program</arg><opt> | --on-stop=</opt><arg>program</arg></p>
		<optdesc><p>
		Execute <arg>program</arg> when playback has ended. Specify the 
		full path to the program, e.g. <file>/usr/bin/logger</file>.
		Executable scripts can be used, but they must have the appropriate shebang
    (<file>#!/bin/sh</file> in the headline.</p>
		<p>If you want shairport-sync to wait until the command has
		completed before continuing, select the <opt>-w</opt> option as well.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>--get-coverart</opt></p>
		<optdesc><p>
		This option requires the <opt>--meta-dir</opt> option to be set, and enables
		shairport-sync to request cover art from the source and to transmit it through
		the metadata pipe.</p>
		<p>Please note that cover art data may be very large, and may place too great a
		burden on your network.
    </p></optdesc>
	  </option>

	  <option>
		<p><opt>-h | --help</opt></p>
		<optdesc><p>
		Print brief help message and exit.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-j</opt></p>
		<optdesc><p>
		Instruct shairport-sync to demonise itself. Unlike the <opt>-d</opt> option, it will not write a
    Process ID (PID) to a file -- it will just (hence the "j") demonise itself.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-k | --kill</opt></p>
		<optdesc><p>
		Kill the shairport-sync daemon and exit. (Requires that the daemon has
		written its PID to an agreed file -- see the <opt>-d</opt> option).
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>--logOutputLevel</opt></p>
		<optdesc><p>
		Use this to log the volume level when the volume is changed. It may be useful if you are trying to
		determine a suitable value for the maximum volume level. Not available as a configuration file setting.
    </p>
    </optdesc>
	  </option>

	  <option>
		<p><opt>-L | --latency=</opt><arg>latency</arg></p>
		<optdesc><p>
		Use this to set the <arg>default latency</arg>, in frames, for audio coming from an unidentified source or from an iTunes Version 9 or earlier source. The standard value for the <arg>default latency</arg> is 88,200 frames, where there are 44,100
    frames to the second.
    </p>
    <p>Please note that this feature is deprecated and will be removed in a future version of shairport-sync.</p>
    </optdesc>
	  </option>

	  <option>
		<p><opt>--meta-dir=</opt><arg>directory</arg></p>
		<optdesc><p>
		Listen for metadata coming from the source and send it, along with metadata from
		shairport-sync itself, to a pipe called <arg>shairport-sync-metadata</arg>
		in the <arg>directory</arg> you specify. If you add the <opt>--get-cover-art</opt> then
		cover art will be sent through the pipe too. See <url href="https://github.com/mikebrady/shairport-sync-metadata-reader"/>
		for a sample metadata reader.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-m </opt><arg>mdnsbackend</arg><opt> | --mdns=</opt><arg>mdnsbackend</arg></p>
		<optdesc><p>
		Force the use of the specified mDNS backend to advertise the
    player on the network. The default is to try all mDNS backends until one
    works.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-o </opt><arg>outputbackend</arg><opt> | --output=</opt><arg>outputbackend</arg></p>
		<optdesc><p>
		Force the use of the specified output backend to play the audio.
    The default is to try the first one.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-p </opt><arg>port</arg><opt> | --port=</opt><arg>port</arg></p>
		<optdesc><p>
		Listen for play requests on <arg>port</arg>. The default is to use port
    5000.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>--password=</opt><arg>secret</arg></p>
		<optdesc><p>
		Require the password <arg>secret</arg> to be able to connect and stream to the service.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-R | --reconnectToOutput</opt></p>
		<optdesc><p>
		Reconnect the shairport-sync daemon to the output device and
    exit. It may take a few seconds to synchronise. (Requires  that
    the daemon has written its PID to an agreed file -- see the <opt>-d</opt>
    option).
		</p>
    <p>Please note that this feature is deprecated and will be removed in a future version of shairport-sync.</p>
		</optdesc>
	  </option>

	  <option>
		<p><opt>-r </opt><arg>threshold</arg><opt> | --resync=</opt><arg>threshold</arg></p>
		<optdesc><p>
		Resynchronise if timings differ by more than <arg>threshold</arg> frames.
    If the output timing differs from the source timing by more than
    the threshold, output will be muted and a full resynchronisation
    will occur. The default threshold is 2,205 frames, i.e. 50
    milliseconds. Specify <opt>0</opt> to disable resynchronisation. This setting is deprecated and will be removed in a future version of shairport-sync.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>--statistics</opt></p>
		<optdesc><p>
		Print some statistics in the standard output, or in the logfile if in daemon mode.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-S </opt><arg>mode</arg><opt> | --stuffing=</opt><arg>mode</arg></p>
		<optdesc><p>
		Stuff the audio stream using the <arg>mode</arg>.  "Stuffing" refers to the
    process of adding or removing frames of audio  to  or  from  the
    stream sent to the output device to keep it exactly in synchrony
    with the player.
    The default mode, <opt>basic</opt>, is normally almost  completely  inaudible.
    The  alternative mode, <opt>soxr</opt>, is even less obtrusive but
    requires much more processing power. For this mode, support for
    libsoxr, the SoX Resampler Library, must be selected when
    shairport-sync is compiled.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-t </opt><arg>timeout</arg><opt> | --timeout=</opt><arg>timeout</arg></p>
		<optdesc><p>
		Exit play mode if the stream disappears for  more  than  <arg>timeout</arg>
    seconds.</p>
    <p>When shairport-sync plays an audio stream, it starts a play
    session and will return a busy signal to any other sources that
    attempt to use it. If the audio stream disappears for longer
    than <arg>timeout</arg> seconds, the play session will be terminated.
    If you  specify a timeout time of <opt>0</opt>,
    shairport-sync will never signal that
    it is busy and will not prevent other sources from "barging in"
    on an existing play session. The default value is 120 seconds.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>--tolerance=</opt><arg>frames</arg></p>
		<optdesc><p>
		Allow playback to be up to <arg>frames</arg> out of exact synchronization before attempting to correct it.
		The default is 88 frames, i.e. 2 ms. The smaller the tolerance, the more likely it is that overcorrection will occur.
		Overcorrection is when more corrections (insertions and deletions) are made than are strictly necessary to keep the stream in sync. Use the <opt>--statistics</opt> option to
		monitor correction levels. Corrections should not greatly exceed net corrections. This setting is deprecated and will be removed in a future version of shairport-sync.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-V | --version</opt></p>
		<optdesc><p>
		Print version information and exit.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-v | --verbose</opt></p>
		<optdesc><p>
		Print debug information. Repeat up to three times to get more detail.
		</p></optdesc>
	  </option>

	  <option>
		<p><opt>-w | --wait-cmd</opt></p>
		<optdesc><p>
		Wait for commands specified using <opt>-B</opt> or <opt>-E</opt>  to  complete  before
    continuing execution.
		</p></optdesc>
	  </option>

  <section name="Audio Backend Options">
  <p>These  command-line options are passed to the chosen audio backend. The audio  backend  options  are
  preceded by a <opt>--</opt> symbol to introduce them and to separate them from any
  program options. In this way, option letters can be used  as  program
  options and also as audio backend options without ambiguity.</p>
  
  <p>In the ALSA backend, audio is sent to an output device
  which you can specify using the <opt>-d</opt> option.
  The output level (the "volume") is controlled using a level control associated with a mixer.
  By default, the mixer is implemented in shairport-sync itself in software.
  To use a hardware level control on a mixer on the sound card, specify the name of the mixer control with the <opt>-c</opt> option.
  If the mixer is not associated with the output device, then you need to specify where the mixer is to be found with the <opt>-m</opt> option.</p>
  </section>
  
  <option>
  <p><opt>-c </opt><arg>controlname</arg></p>
  <optdesc><p>
  Use the level control called <arg>controlname</arg> on the hardware mixer for
  controlling volume.
  This is needed if the mixer type is specified, using the <opt>-t</opt> option,
  to be <opt>hardware</opt>. There is no default.
  </p></optdesc>
  </option>

  <option>
  <p><opt>-d </opt><arg>device</arg></p>
  <optdesc><p>
  Use the specified output <arg>device</arg>. You may specify a card, e.g. <opt>hw:0</opt>, in
  which case the default output device on the card will be chosen.
  Alternatively, you can specify a specific device on a card, e.g. <opt>hw:0,0</opt>.
  The default is the device named <opt>default</opt>.
  </p></optdesc>
  </option>

  <option>
  <p><opt>-m </opt><arg>mixer</arg></p>
  <optdesc><p>
  Use the specified hardware <arg>mixer</arg> for volume control. Use this to specify where
  the mixer is to be found. For example, if the mixer is associated with a card,
  as is often the case, specify the card, e.g. <opt>hw:0</opt>.
  If (unusually) the mixer is associated with a specific device on a card,
  specify the device, e.g. <opt>hw:0,1</opt>.
  The default is the device named in the <opt>-d</opt> option,
  if given, or the device named <opt>default</opt>.
  </p></optdesc>
  </option>

  <option>
  <p><opt>-t </opt><arg>devicetype</arg></p>
  <optdesc>
  <p>
  This option is deprecated and is ignored. For your information, its functionality has been automatically incorporated in the -c option
  -- if you specify a mixer name with the -c option, it is assumed that the mixer is implemented in hardware.
  </p></optdesc>
  </option>
	</options>

  <section name="Examples">
  <p>Here is a slightly contrived example:</p>
      <cmd>shairport-sync  <opt>-d</opt>
      <opt>-a "Joe's Stereo"</opt>
      <opt>-S soxr</opt>
      <opt>--</opt>
      <opt>-d hw:1,0</opt>
      <opt>-m hw:1</opt>
      <opt>-c PCM</opt>      
      </cmd>
  <p>The program will run in daemon mode ( <opt>-d</opt> ), will be  visible  as
  "Joe's Stereo" ( <opt>-a "Joe's Stereo"</opt> ) and will use the SoX Resampler
  Library-based stuffing ( <opt>-S soxr</opt> ).
  The audio backend options following the <opt>--</opt> separator specify
  that  the  audio will be output on output 0 of soundcard 1 (
  <opt>-d hw:1,0</opt> ) and will take advantage of the same sound card's mixer ( <opt>-m hw:1</opt> )
  using the level control named "PCM" ( <opt>-c "PCM"</opt> ).
  </p>
  <p>The example above is slightly contrived in order to show the use of the <opt>-m</opt> option.
  Typically, output 0 is the default output of a card,
  so the output device could be written <opt>-d hw:1</opt> and
  then the mixer option would be unnecessary, giving the following, simpler, command:</p>
       <cmd>shairport-sync  <opt>-d</opt>
      <opt>-a "Joe's Stereo"</opt>
      <opt>-S soxr</opt>
      <opt>--</opt>
      <opt>-d hw:1</opt>
      <opt>-c PCM</opt>      
      </cmd>
 
  </section>

	<section name="Credits">
	<p>Mike Brady developed shairport-sync from the original Shairport by James Laird.</p>
	<p>shairport-sync can be found at  <url href="https://github.com/mikebrady/shairport-sync."/></p>
  <p>Shairport can be found at <url href="https://github.com/abrasive/shairport."/></p>
	</section>

	<section name="Comments">
	  <p>This man page was written using <manref name="xml2man" section="1"
		  href="http://masqmail.cx/xml2man/"/> by Oliver Kurth.</p>
	</section>

  </manpage>