path: root/usr.bin/aucat/aucat.1

.\"	$OpenBSD: aucat.1,v 1.53 2009/05/09 09:50:40 ratchov Exp $
.\"
.\" Copyright (c) 2006 Alexandre Ratchov <alex@caoua.org>
.\"
.\" Permission to use, copy, modify, and distribute this software for any
.\" purpose with or without fee is hereby granted, provided that the above
.\" copyright notice and this permission notice appear in all copies.
.\"
.\" THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
.\" WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
.\" MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
.\" ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
.\" WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
.\" ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
.\" OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
.\"
.Dd $Mdocdate: May 9 2009 $
.Dt AUCAT 1
.Os
.Sh NAME
.Nm aucat
.Nd audio server and stream manipulation tool
.Sh SYNOPSIS
.Nm aucat
.Bk -words
.Op Fl lnu
.Op Fl b Ar nframes
.Op Fl C Ar min : Ns Ar max
.Op Fl c Ar min : Ns Ar max
.Op Fl e Ar enc
.Op Fl f Ar device
.Op Fl h Ar fmt
.Op Fl i Ar file
.Op Fl m Ar mode
.Op Fl o Ar file
.Op Fl r Ar rate
.Op Fl s Ar socket
.Op Fl v Ar volume
.Op Fl x Ar policy
.Ek
.Sh DESCRIPTION
The
.Nm
utility can record one input stream
and store it on multiple destination files,
doing the necessary conversions on the fly.
It can play, convert, and mix multiple input files simultaneously,
and can also act as an audio server.
.Nm
also has a legacy mode that works like previous versions of
.Nm ,
which does not convert on the fly and supports playback of .au files.
.Pp
The options are as follows:
.Bl -tag -width Ds
.It Fl b Ar nframes
The buffer size in frames.
A frame consists of one sample for each channel in the stream.
This is the number of frames that will be buffered before being played
and thus controls the playback latency.
.It Xo
.Fl C Ar min : Ns Ar max ,
.Fl c Ar min : Ns Ar max
.Xc
The range of channel numbers on the record or playback stream, respectively.
The default is 0:1, i.e. stereo.
.It Fl e Ar enc
Encoding of the playback or recording stream (see below).
The default is signed, 16-bit, native byte order.
.It Fl f Ar device
The
.Xr audio 4
device or
.Nm
socket to use for playing and/or recording.
.It Fl h Ar fmt
File format of the playback or record stream (see below).
The default is auto.
.It Fl i Ar file
Add this file to the list of files to play.
If the option argument is
.Sq -
then standard input will be used.
.It Fl l
Listen for incoming connections on Unix domain sockets.
This allows clients to use
.Nm
instead of the regular
.Xr audio 4
device for audio input and output
in order to share the physical device with other clients.
The default socket name is
.Pa default
but other names can be used with the
.Fl s
option.
.It Fl m Ar mode
Set the server mode.
Valid modes are
.Ar play ,
.Ar rec ,
and
.Ar duplex ,
for play-only, record-only, and full-duplex, respectively.
The default is
.Ar duplex .
.It Fl n
Loopback mode.
Instead of using an audio device, send input streams
to the output, processing them on the fly.
This mode is useful to mix, demultiplex, resample or reencode
audio files offline.
.It Fl o Ar file
Add this file to the list of files in which to store recorded samples.
If the option argument is
.Sq -
then standard output will be used.
.It Fl r Ar rate
Sample rate in Hertz of the playback or record stream.
The default is 44100Hz.
.It Fl s Ar socket
Add the name
.Ar socket
to the list of sockets to listen on.
.Ar socket
cannot contain '/'.
Meaningful in server mode only.
.It Fl u
Normally
.Nm
tries to automatically determine the optimal parameters for the audio device;
if this option is specified,
it will instead use the parameters specified by the
.Fl Ccer
options.
.It Fl v Ar volume
Software volume attenuation of the playback stream.
The value must be between 1 and 127,
corresponding to \-42dB and \-0dB attenuation.
In server mode, clients inherit this parameter.
Reducing the volume in advance reduces a client's dynamic range,
but allows client volume to stay independent from the number
of clients as long as their number is small enough.
A good compromise is to use \-4dB attenuation (12 volume units)
for each additional client expected
(115 if 2 clients are expected, 103 for 3 clients, and so on).
.It Fl x Ar policy
Action when the output stream cannot accept
recorded data fast enough or the input stream
cannot provide data to play fast enough.
If the policy
is
.Dq ignore
(the default) then samples that cannot be written are discarded
and samples that cannot be read are replaced by silence.
If the policy is
.Dq sync
then recorded samples are discarded, but the same amount of silence will be written
once the stream is unblocked, in order to reach the right position in time.
Similarly silence is played, but the same amount of samples will be discarded
once the stream is unblocked.
If the policy is
.Dq error
then the stream is closed permanently.
.El
.Pp
If
.Nm
is sent
.Dv SIGHUP ,
.Dv SIGINT
or
.Dv SIGTERM ,
it terminates recording to files.
If sent
.Dv SIGUSR1
or
.Dv SIGUSR2 ,
it increases or decreases debug level, respectively.
.Pp
Settings for input files
.Pq Fl i ,
output files
.Pq Fl o ,
and sockets
.Pq Fl s
can be changed using the
.Fl Ccehrvx
options.
The last
.Fl Ccehrvx
options specified before an
.Fl i ,
.Fl o ,
or
.Fl s
are applied to the corresponding file.
.Pp
Settings for the audio device
can be changed using the
.Fl Ccer
options.
They apply to the audio device only if the
.Fl u
option is given as well.
The last
.Fl Ccer
option specified before an
.Fl f
is applied to
.Ar device .
.Pp
If no audio device
.Pq Fl f
is specified,
settings are applied as if
the default device is specified as the last argument.
If no sockets
.Pq Fl s
are specified
settings are applied as if
the default socket is specified as the last argument.
.Pp
File formats are specified using the
.Fl h
option.
The following file formats are supported:
.Bl -tag -width s32lexxx -offset -indent
.It raw
Headerless file.
This format is recommended since it has no limitations.
.It wav
Microsoft WAVE file format.
There are limitations inherent to the file format itself:
not all encodings are supported,
file sizes are limited to 2GB,
and the file must support the
.Xr lseek 2
operation (e.g. pipes do not support it).
.It auto
Try to guess, depending on the file name.
.El
.Pp
Encodings are specified using the
.Fl e
option.
The following encodings are supported:
.Pp
.Bl -tag -width s32lexxx -offset -indent -compact
.It s8
signed 8-bit
.It u8
unsigned 8-bit
.It s16le
signed 16-bit, little endian
.It u16le
unsigned 16-bit, little endian
.It s16be
signed 16-bit, big endian
.It u16be
unsigned 16-bit, big endian
.It s24le
signed 24-bit, stored in 4 bytes, little endian
.It u24le
unsigned 24-bit, stored in 4 bytes, little endian
.It s24be
signed 24-bit, stored in 4 bytes, big endian
.It u24be
unsigned 24-bit, stored in 4 bytes, big endian
.It s32le
signed 32-bit, little endian
.It u32le
unsigned 32-bit, little endian
.It s32be
signed 32-bit, big endian
.It u32be
unsigned 32-bit, big endian
.It s24le3
signed 24-bit, packed in 3 bytes, little endian
.It u24le3
unsigned 24-bit, packed in 3 bytes, big endian
.It s24be3
signed 24-bit, packed in 3 bytes, little endian
.It u24be3
unsigned 24-bit, packed in 3 bytes, big endian
.It s20le3
signed 20-bit, packed in 3 bytes, little endian
.It u20le3
unsigned 20-bit, packed in 3 bytes, big endian
.It s20be3
signed 20-bit, packed in 3 bytes, little endian
.It u20be3
unsigned 20-bit, packed in 3 bytes, big endian
.It s18le3
signed 18-bit, packed in 3 bytes, little endian
.It u18le3
unsigned 18-bit, packed in 3 bytes, big endian
.It s18be3
signed 18-bit, packed in 3 bytes, little endian
.It u18be3
unsigned 18-bit, packed in 3 bytes, big endian
.El
.Sh SERVER MODE
.Nm
can be used in server mode
.Pq Fl l
to overcome hardware limitations and allow applications
to run on fixed sample rate devices or on devices
supporting only unusual encodings.
It is generally not desirable to have multiple
instances of
.Nm
running in server mode,
so it is good practice to start it thus:
.Bd -literal -offset indent
$ pgrep -x aucat || aucat -l
.Ed
.Pp
This also ensures privacy by preventing
other users from accessing the audio system.
On multi-user machines
.Nm
should be killed when no longer in use to make audio resources
available again to others:
.Bd -literal -offset indent
$ pkill -x aucat
.Ed
.Pp
Certain applications, such as synthesis software,
require a low latency audio setup.
To reduce the probability of buffer underruns or overruns,
the
.Xr renice 8
command can be used to give a higher priority to the
.Nm
process.
Superuser privileges are required.
For example:
.Bd -literal -offset indent
$ aucat -b 3500 -l
$ sudo renice -n -20 -p `pgrep -x aucat`
.Ed
.Sh LEGACY MODE
If neither
.Fl i
nor
.Fl o
are specified,
.Nm
will run in legacy mode, and won't convert sample formats or sampling rates.
In legacy mode, all options except
.Fl f
are ignored, and all other arguments are assumed to be names of files.
In legacy mode
.Nm
reads files sequentially, and writes them to the specified device.
If a Sun .au header is detected it is skipped over and not copied to
the audio device.
.Nm
will attempt to parse the format, number of channels and sample rate
from Sun .au file headers.
However, only alaw and ulaw formats are supported for .au files.
Other formats will be interpreted as ulaw.
If a Microsoft .wav header (RIFF) is detected it is interpreted
to select the right audio encoding for playback and the data chunk of the
file is copied to the audio device.
If the device does not support the encoding,
.Nm
will exit with an error.
.Sh ENVIRONMENT
.Bl -tag -width "AUDIODEVICE" -compact
.It Ev AUCAT_DEBUG
The debug level:
may be a value between 0 and 4.
.It Ev AUDIODEVICE
Name of the
.Nm
socket to connect to, or path to the
.Xr audio 4
device to use if the
.Fl f
option is not specified.
.El
.Sh EXAMPLES
The following will mix and play two stereo streams,
the first at 48kHz and the second at 44.1kHz:
.Bd -literal -offset indent
$ aucat -r 48000 -i file1.raw -r 44100 -i file2.raw
.Ed
.Pp
The following will record channels 2 and 3 into one stereo file and
channels 6 and 7 into another stereo file using a 96kHz sampling rate for
both:
.Bd -literal -offset indent
$ aucat -r 96000 -C 2:3 -o file1.raw -C 6:7 -o file2.raw
.Ed
.Pp
The following will split a stereo file into two mono files:
.Bd -literal -offset indent
$ aucat -n -i stereo.wav -C 0:0 -o left.wav -C 1:1 -o right.wav
.Ed
.Pp
The following will start
.Nm
in server mode using default parameters, but will create an
additional socket for output to channels 2:3 only (rear speakers
on most cards):
.Bd -literal -offset indent
$ aucat -l -s default -c 2:3 -s rear
.Ed
.Pp
The following will start
.Nm
in server mode creating the default socket with low volume and
an additional socket for high volume output:
.Bd -literal -offset indent
$ aucat -l -v 65 -s default -v 127 -s max
.Ed
.Sh SEE ALSO
.Xr audioctl 1 ,
.Xr cdio 1 ,
.Xr mixerctl 1 ,
.Xr audio 4
.Sh BUGS
The
.Nm
utility assumes non-blocking I/O for input and output streams.
It will not work reliably on files that may block
(ordinary files block, pipes don't).
.Pp
Resampling is low quality; down-sampling especially should be avoided
when recording.
.Pp
Processing is done using 16-bit arithmetic,
thus samples with more than 16 bits are rounded.
16 bits (i.e. 97dB dynamic) are largely enough for most applications though.