1
0
mirror of https://github.com/mpv-player/mpv synced 2024-12-27 17:42:17 +00:00
mpv/DOCS/man/af.rst

223 lines
9.1 KiB
ReStructuredText
Raw Normal View History

AUDIO FILTERS
=============
Audio filters allow you to modify the audio stream and its properties. The
syntax is:
``--af=...``
Setup a chain of audio filters. See ``--vf`` (`VIDEO FILTERS`_) for the
full syntax.
2013-07-08 16:02:14 +00:00
.. note::
To get a full list of available audio filters, see ``--af=help``.
2015-09-11 21:48:47 +00:00
Also, keep in mind that most actual filters are available via the ``lavfi``
wrapper, which gives you access to most of libavfilter's filters. This
includes all filters that have been ported from MPlayer to libavfilter.
The ``--vf`` description describes how libavfilter can be used and how to
workaround deprecated mpv filters.
See ``--vf`` group of options for info on how ``--af-defaults``, ``--af-add``,
``--af-pre``, ``--af-del``, ``--af-clr``, and possibly others work.
Available filters are:
``lavcac3enc[=options]``
Encode multi-channel audio to AC-3 at runtime using libavcodec. Supports
16-bit native-endian input format, maximum 6 channels. The output is
big-endian when outputting a raw AC-3 stream, native-endian when
2013-07-08 16:02:14 +00:00
outputting to S/PDIF. If the input sample rate is not 48 kHz, 44.1 kHz or
32 kHz, it will be resampled to 48 kHz.
``tospdif=<yes|no>``
Output raw AC-3 stream if ``no``, output to S/PDIF for
2014-09-01 02:25:57 +00:00
pass-through if ``yes`` (default).
``bitrate=<rate>``
The bitrate use for the AC-3 stream. Set it to 384 to get 384 kbps.
The default is 640. Some receivers might not be able to handle this.
Valid values: 32, 40, 48, 56, 64, 80, 96, 112, 128,
160, 192, 224, 256, 320, 384, 448, 512, 576, 640.
The special value ``auto`` selects a default bitrate based on the
input channel number:
:1ch: 96
:2ch: 192
:3ch: 224
:4ch: 384
:5ch: 448
:6ch: 448
``minch=<n>``
If the input channel number is less than ``<minch>``, the filter will
detach itself (default: 3).
``encoder=<name>``
Select the libavcodec encoder used. Currently, this should be an AC-3
encoder, and using another codec will fail horribly.
``format=format:srate:channels:out-srate:out-channels``
2015-04-15 10:48:05 +00:00
Does not do any format conversion itself. Rather, it may cause the
filter system to insert necessary conversion filters before or after this
filter if needed. It is primarily useful for controlling the audio format
going into other filters. To specify the format for audio output, see
``--audio-format``, ``--audio-samplerate``, and ``--audio-channels``. This
filter is able to force a particular format, whereas ``--audio-*``
may be overridden by the ao based on output compatibility.
All parameters are optional. The first 3 parameters restrict what the filter
2015-04-15 10:48:05 +00:00
accepts as input. They will therefore cause conversion filters to be
inserted before this one. The ``out-`` parameters tell the filters or audio
outputs following this filter how to interpret the data without actually
doing a conversion. Setting these will probably just break things unless you
really know you want this for some reason, such as testing or dealing with
broken media.
audio/filter: split af_format into separate filters, rename af_force af_format is the old audio conversion filter. It could do all possible conversions supported by the audio chain. However, ever since the addition of af_lavrresample, most conversions are done by libav/swresample, and af_format is used as fallback. Separate out the fallback cases and remove af_format. af_convert24 does 24 bit <-> 32 bit conversions, while af_convertsignendian does sign and endian conversions. Maybe the way the conversions are split sounds a bit odd. But the former changes the size of the audio data, while the latter is fully in-place, so there's at least different buffer management. This requires a quite complicated algorithm to make sure all these "partial" conversion filters can actually get from one format to another. E.g. s24le->s32be always requires convertsignendian and convert24, but af.c has no idea what the intermediate format should be. So I added a graph search (trying every possible format and filter) to determine required format and filter. When I wrote this, it seemed this was still better than messing everything into af_lavrresample, but maybe this is overkill and I'll change my opinion. For now, it seems nice to get rid of af_format though. The AC3->IEC61937 conversion isn't supported anymore, but I don't think this is needed anywhere. Most AOs test all formats explicitly, or use the AF_FORMAT_IS_IEC61937() macro (which includes AC3). One positive consequence of this change is that conversions always include dithering (done by libav/swresample), instead of possibly going through af_format, which doesn't do anything fancy. Rename af_force to af_format. It's essentially compatible with command line uses of af_format. We retain a compatibility alias for af_force.
2013-10-21 23:20:43 +00:00
``<format>``
Force conversion to this format. Use ``--af=format=format=help`` to get
a list of valid formats.
audio/filter: split af_format into separate filters, rename af_force af_format is the old audio conversion filter. It could do all possible conversions supported by the audio chain. However, ever since the addition of af_lavrresample, most conversions are done by libav/swresample, and af_format is used as fallback. Separate out the fallback cases and remove af_format. af_convert24 does 24 bit <-> 32 bit conversions, while af_convertsignendian does sign and endian conversions. Maybe the way the conversions are split sounds a bit odd. But the former changes the size of the audio data, while the latter is fully in-place, so there's at least different buffer management. This requires a quite complicated algorithm to make sure all these "partial" conversion filters can actually get from one format to another. E.g. s24le->s32be always requires convertsignendian and convert24, but af.c has no idea what the intermediate format should be. So I added a graph search (trying every possible format and filter) to determine required format and filter. When I wrote this, it seemed this was still better than messing everything into af_lavrresample, but maybe this is overkill and I'll change my opinion. For now, it seems nice to get rid of af_format though. The AC3->IEC61937 conversion isn't supported anymore, but I don't think this is needed anywhere. Most AOs test all formats explicitly, or use the AF_FORMAT_IS_IEC61937() macro (which includes AC3). One positive consequence of this change is that conversions always include dithering (done by libav/swresample), instead of possibly going through af_format, which doesn't do anything fancy. Rename af_force to af_format. It's essentially compatible with command line uses of af_format. We retain a compatibility alias for af_force.
2013-10-21 23:20:43 +00:00
``<srate>``
Force conversion to a specific sample rate. The rate is an integer,
48000 for example.
audio/filter: split af_format into separate filters, rename af_force af_format is the old audio conversion filter. It could do all possible conversions supported by the audio chain. However, ever since the addition of af_lavrresample, most conversions are done by libav/swresample, and af_format is used as fallback. Separate out the fallback cases and remove af_format. af_convert24 does 24 bit <-> 32 bit conversions, while af_convertsignendian does sign and endian conversions. Maybe the way the conversions are split sounds a bit odd. But the former changes the size of the audio data, while the latter is fully in-place, so there's at least different buffer management. This requires a quite complicated algorithm to make sure all these "partial" conversion filters can actually get from one format to another. E.g. s24le->s32be always requires convertsignendian and convert24, but af.c has no idea what the intermediate format should be. So I added a graph search (trying every possible format and filter) to determine required format and filter. When I wrote this, it seemed this was still better than messing everything into af_lavrresample, but maybe this is overkill and I'll change my opinion. For now, it seems nice to get rid of af_format though. The AC3->IEC61937 conversion isn't supported anymore, but I don't think this is needed anywhere. Most AOs test all formats explicitly, or use the AF_FORMAT_IS_IEC61937() macro (which includes AC3). One positive consequence of this change is that conversions always include dithering (done by libav/swresample), instead of possibly going through af_format, which doesn't do anything fancy. Rename af_force to af_format. It's essentially compatible with command line uses of af_format. We retain a compatibility alias for af_force.
2013-10-21 23:20:43 +00:00
``<channels>``
Force mixing to a specific channel layout. See ``--audio-channels`` option
for possible values.
2013-07-08 16:02:14 +00:00
``<out-srate>``
2013-07-08 16:02:14 +00:00
``<out-channels>``
2015-04-15 10:48:05 +00:00
*NOTE*: this filter used to be named ``force``. The old ``format`` filter
used to do conversion itself, unlike this one which lets the filter system
handle the conversion.
audio/filter: split af_format into separate filters, rename af_force af_format is the old audio conversion filter. It could do all possible conversions supported by the audio chain. However, ever since the addition of af_lavrresample, most conversions are done by libav/swresample, and af_format is used as fallback. Separate out the fallback cases and remove af_format. af_convert24 does 24 bit <-> 32 bit conversions, while af_convertsignendian does sign and endian conversions. Maybe the way the conversions are split sounds a bit odd. But the former changes the size of the audio data, while the latter is fully in-place, so there's at least different buffer management. This requires a quite complicated algorithm to make sure all these "partial" conversion filters can actually get from one format to another. E.g. s24le->s32be always requires convertsignendian and convert24, but af.c has no idea what the intermediate format should be. So I added a graph search (trying every possible format and filter) to determine required format and filter. When I wrote this, it seemed this was still better than messing everything into af_lavrresample, but maybe this is overkill and I'll change my opinion. For now, it seems nice to get rid of af_format though. The AC3->IEC61937 conversion isn't supported anymore, but I don't think this is needed anywhere. Most AOs test all formats explicitly, or use the AF_FORMAT_IS_IEC61937() macro (which includes AC3). One positive consequence of this change is that conversions always include dithering (done by libav/swresample), instead of possibly going through af_format, which doesn't do anything fancy. Rename af_force to af_format. It's essentially compatible with command line uses of af_format. We retain a compatibility alias for af_force.
2013-10-21 23:20:43 +00:00
2013-07-08 16:02:14 +00:00
``scaletempo[=option1:option2:...]``
Scales audio tempo without altering pitch, optionally synced to playback
speed (default).
This works by playing 'stride' ms of audio at normal speed then consuming
'stride*scale' ms of input audio. It pieces the strides together by
blending 'overlap'% of stride with audio following the previous stride. It
optionally performs a short statistical analysis on the next 'search' ms
of audio to determine the best overlap position.
2013-07-08 16:02:14 +00:00
``scale=<amount>``
Nominal amount to scale tempo. Scales this amount in addition to
speed. (default: 1.0)
2013-07-08 16:02:14 +00:00
``stride=<amount>``
Length in milliseconds to output each stride. Too high of a value will
cause noticeable skips at high scale amounts and an echo at low scale
amounts. Very low values will alter pitch. Increasing improves
performance. (default: 60)
2013-07-08 16:02:14 +00:00
``overlap=<percent>``
Percentage of stride to overlap. Decreasing improves performance.
(default: .20)
2013-07-08 16:02:14 +00:00
``search=<amount>``
Length in milliseconds to search for best overlap position. Decreasing
improves performance greatly. On slow systems, you will probably want
to set this very low. (default: 14)
2013-07-08 16:02:14 +00:00
``speed=<tempo|pitch|both|none>``
Set response to speed change.
tempo
Scale tempo in sync with speed (default).
pitch
Reverses effect of filter. Scales pitch without altering tempo.
2015-03-15 14:18:11 +00:00
Add this to your ``input.conf`` to step by musical semi-tones::
[ multiply speed 0.9438743126816935
] multiply speed 1.059463094352953
2013-07-08 16:02:14 +00:00
.. warning::
Loses sync with video.
both
Scale both tempo and pitch.
none
Ignore speed changes.
2013-07-08 16:02:14 +00:00
.. admonition:: Examples
2013-07-08 16:02:14 +00:00
``mpv --af=scaletempo --speed=1.2 media.ogg``
Would play media at 1.2x normal speed, with audio at normal
pitch. Changing playback speed would change audio tempo to match.
2013-07-08 16:02:14 +00:00
``mpv --af=scaletempo=scale=1.2:speed=none --speed=1.2 media.ogg``
Would play media at 1.2x normal speed, with audio at normal
pitch, but changing playback speed would have no effect on audio
tempo.
2013-07-08 16:02:14 +00:00
``mpv --af=scaletempo=stride=30:overlap=.50:search=10 media.ogg``
2014-09-01 02:25:57 +00:00
Would tweak the quality and performance parameters.
2013-07-08 16:02:14 +00:00
``mpv --af=scaletempo=scale=1.2:speed=pitch audio.ogg``
Would play media at 1.2x normal speed, with audio at normal pitch.
Changing playback speed would change pitch, leaving audio tempo at
1.2x.
``rubberband``
High quality pitch correction with librubberband. This can be used in place
of ``scaletempo``, and will be used to adjust audio pitch when playing
at speed different from normal. It can also be used to adjust audio pitch
without changing playback speed.
``<pitch-scale>``
Sets the pitch scaling factor. Frequencies are multiplied by this value.
This filter has a number of additional sub-options. You can list them with
``mpv --af=rubberband=help``. This will also show the default values
for each option. The options are not documented here, because they are
merely passed to librubberband. Look at the librubberband documentation
to learn what each option does:
http://breakfastquay.com/rubberband/code-doc/classRubberBand_1_1RubberBandStretcher.html
(The mapping of the mpv rubberband filter sub-option names and values to
those of librubberband follows a simple pattern: ``"Option" + Name + Value``.)
This filter supports the following ``af-command`` commands:
``set-pitch``
Set the ``<pitch-scale>`` argument dynamically. This can be used to
change the playback pitch at runtime. Note that speed is controlled
using the standard ``speed`` property, not ``af-command``.
``multiply-pitch <factor>``
Multiply the current value of ``<pitch-scale>`` dynamically. For
example: 0.5 to go down by an octave, 1.5 to go up by a perfect fifth.
If you want to go up or down by semi-tones, use 1.059463094352953 and
0.9438743126816935
2013-07-08 16:02:14 +00:00
``lavfi=graph``
2014-09-01 02:25:57 +00:00
Filter audio using FFmpeg's libavfilter.
2013-07-08 16:02:14 +00:00
``<graph>``
Libavfilter graph. See ``lavfi`` video filter for details - the graph
syntax is the same.
2013-07-08 16:02:14 +00:00
.. warning::
Don't forget to quote libavfilter graphs as described in the lavfi
video filter section.
``o=<string>``
AVOptions.
``fix-pts=<yes|no>``
Determine PTS based on sample count (default: no). If this is enabled,
the player won't rely on libavfilter passing through PTS accurately.
Instead, it pass a sample count as PTS to libavfilter, and compute the
PTS used by mpv based on that and the input PTS. This helps with filters
which output a recomputed PTS instead of the original PTS (including
filters which require the PTS to start at 0). mpv normally expects
filters to not touch the PTS (or only to the extent of changing frame
boundaries), so this is not the default, but it will be needed to use
broken filters. In practice, these broken filters will either cause slow
A/V desync over time (with some files), or break playback completely if
you seek or start playback from the middle of a file.