1 @chapter Filtering Introduction
2 @c man begin FILTERING INTRODUCTION
4 Filtering in FFmpeg is enabled through the libavfilter library.
6 In libavfilter, a filter can have multiple inputs and multiple
8 To illustrate the sorts of things that are possible, we consider the
13 input --> split ---------------------> overlay --> output
16 +-----> crop --> vflip -------+
19 This filtergraph splits the input stream in two streams, then sends one
20 stream through the crop filter and the vflip filter, before merging it
21 back with the other stream by overlaying it on top. You can use the
22 following command to achieve this:
25 ffmpeg -i INPUT -vf "split [main][tmp]; [tmp] crop=iw:ih/2:0:0, vflip [flip]; [main][flip] overlay=0:H/2" OUTPUT
28 The result will be that the top half of the video is mirrored
29 onto the bottom half of the output video.
31 Filters in the same linear chain are separated by commas, and distinct
32 linear chains of filters are separated by semicolons. In our example,
33 @var{crop,vflip} are in one linear chain, @var{split} and
34 @var{overlay} are separately in another. The points where the linear
35 chains join are labelled by names enclosed in square brackets. In the
36 example, the split filter generates two outputs that are associated to
37 the labels @var{[main]} and @var{[tmp]}.
39 The stream sent to the second output of @var{split}, labelled as
40 @var{[tmp]}, is processed through the @var{crop} filter, which crops
41 away the lower half part of the video, and then vertically flipped. The
42 @var{overlay} filter takes in input the first unchanged output of the
43 split filter (which was labelled as @var{[main]}), and overlay on its
44 lower half the output generated by the @var{crop,vflip} filterchain.
46 Some filters take in input a list of parameters: they are specified
47 after the filter name and an equal sign, and are separated from each other
50 There exist so-called @var{source filters} that do not have an
51 audio/video input, and @var{sink filters} that will not have audio/video
54 @c man end FILTERING INTRODUCTION
57 @c man begin GRAPH2DOT
59 The @file{graph2dot} program included in the FFmpeg @file{tools}
60 directory can be used to parse a filtergraph description and issue a
61 corresponding textual representation in the dot language.
68 to see how to use @file{graph2dot}.
70 You can then pass the dot description to the @file{dot} program (from
71 the graphviz suite of programs) and obtain a graphical representation
74 For example the sequence of commands:
76 echo @var{GRAPH_DESCRIPTION} | \
77 tools/graph2dot -o graph.tmp && \
78 dot -Tpng graph.tmp -o graph.png && \
82 can be used to create and display an image representing the graph
83 described by the @var{GRAPH_DESCRIPTION} string. Note that this string must be
84 a complete self-contained graph, with its inputs and outputs explicitly defined.
85 For example if your command line is of the form:
87 ffmpeg -i infile -vf scale=640:360 outfile
89 your @var{GRAPH_DESCRIPTION} string will need to be of the form:
91 nullsrc,scale=640:360,nullsink
93 you may also need to set the @var{nullsrc} parameters and add a @var{format}
94 filter in order to simulate a specific input file.
98 @chapter Filtergraph description
99 @c man begin FILTERGRAPH DESCRIPTION
101 A filtergraph is a directed graph of connected filters. It can contain
102 cycles, and there can be multiple links between a pair of
103 filters. Each link has one input pad on one side connecting it to one
104 filter from which it takes its input, and one output pad on the other
105 side connecting it to one filter accepting its output.
107 Each filter in a filtergraph is an instance of a filter class
108 registered in the application, which defines the features and the
109 number of input and output pads of the filter.
111 A filter with no input pads is called a "source", and a filter with no
112 output pads is called a "sink".
114 @anchor{Filtergraph syntax}
115 @section Filtergraph syntax
117 A filtergraph has a textual representation, which is recognized by the
118 @option{-filter}/@option{-vf}/@option{-af} and
119 @option{-filter_complex} options in @command{ffmpeg} and
120 @option{-vf}/@option{-af} in @command{ffplay}, and by the
121 @code{avfilter_graph_parse_ptr()} function defined in
122 @file{libavfilter/avfilter.h}.
124 A filterchain consists of a sequence of connected filters, each one
125 connected to the previous one in the sequence. A filterchain is
126 represented by a list of ","-separated filter descriptions.
128 A filtergraph consists of a sequence of filterchains. A sequence of
129 filterchains is represented by a list of ";"-separated filterchain
132 A filter is represented by a string of the form:
133 [@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
135 @var{filter_name} is the name of the filter class of which the
136 described filter is an instance of, and has to be the name of one of
137 the filter classes registered in the program.
138 The name of the filter class is optionally followed by a string
141 @var{arguments} is a string which contains the parameters used to
142 initialize the filter instance. It may have one of two forms:
146 A ':'-separated list of @var{key=value} pairs.
149 A ':'-separated list of @var{value}. In this case, the keys are assumed to be
150 the option names in the order they are declared. E.g. the @code{fade} filter
151 declares three options in this order -- @option{type}, @option{start_frame} and
152 @option{nb_frames}. Then the parameter list @var{in:0:30} means that the value
153 @var{in} is assigned to the option @option{type}, @var{0} to
154 @option{start_frame} and @var{30} to @option{nb_frames}.
157 A ':'-separated list of mixed direct @var{value} and long @var{key=value}
158 pairs. The direct @var{value} must precede the @var{key=value} pairs, and
159 follow the same constraints order of the previous point. The following
160 @var{key=value} pairs can be set in any preferred order.
164 If the option value itself is a list of items (e.g. the @code{format} filter
165 takes a list of pixel formats), the items in the list are usually separated by
168 The list of arguments can be quoted using the character @samp{'} as initial
169 and ending mark, and the character @samp{\} for escaping the characters
170 within the quoted text; otherwise the argument string is considered
171 terminated when the next special character (belonging to the set
172 @samp{[]=;,}) is encountered.
174 The name and arguments of the filter are optionally preceded and
175 followed by a list of link labels.
176 A link label allows one to name a link and associate it to a filter output
177 or input pad. The preceding labels @var{in_link_1}
178 ... @var{in_link_N}, are associated to the filter input pads,
179 the following labels @var{out_link_1} ... @var{out_link_M}, are
180 associated to the output pads.
182 When two link labels with the same name are found in the
183 filtergraph, a link between the corresponding input and output pad is
186 If an output pad is not labelled, it is linked by default to the first
187 unlabelled input pad of the next filter in the filterchain.
188 For example in the filterchain
190 nullsrc, split[L1], [L2]overlay, nullsink
192 the split filter instance has two output pads, and the overlay filter
193 instance two input pads. The first output pad of split is labelled
194 "L1", the first input pad of overlay is labelled "L2", and the second
195 output pad of split is linked to the second input pad of overlay,
196 which are both unlabelled.
198 In a filter description, if the input label of the first filter is not
199 specified, "in" is assumed; if the output label of the last filter is not
200 specified, "out" is assumed.
202 In a complete filterchain all the unlabelled filter input and output
203 pads must be connected. A filtergraph is considered valid if all the
204 filter input and output pads of all the filterchains are connected.
206 Libavfilter will automatically insert @ref{scale} filters where format
207 conversion is required. It is possible to specify swscale flags
208 for those automatically inserted scalers by prepending
209 @code{sws_flags=@var{flags};}
210 to the filtergraph description.
212 Here is a BNF description of the filtergraph syntax:
214 @var{NAME} ::= sequence of alphanumeric characters and '_'
215 @var{LINKLABEL} ::= "[" @var{NAME} "]"
216 @var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
217 @var{FILTER_ARGUMENTS} ::= sequence of chars (possibly quoted)
218 @var{FILTER} ::= [@var{LINKLABELS}] @var{NAME} ["=" @var{FILTER_ARGUMENTS}] [@var{LINKLABELS}]
219 @var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
220 @var{FILTERGRAPH} ::= [sws_flags=@var{flags};] @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
223 @section Notes on filtergraph escaping
225 Filtergraph description composition entails several levels of
226 escaping. See @ref{quoting_and_escaping,,the "Quoting and escaping"
227 section in the ffmpeg-utils(1) manual,ffmpeg-utils} for more
228 information about the employed escaping procedure.
230 A first level escaping affects the content of each filter option
231 value, which may contain the special character @code{:} used to
232 separate values, or one of the escaping characters @code{\'}.
234 A second level escaping affects the whole filter description, which
235 may contain the escaping characters @code{\'} or the special
236 characters @code{[],;} used by the filtergraph description.
238 Finally, when you specify a filtergraph on a shell commandline, you
239 need to perform a third level escaping for the shell special
240 characters contained within it.
242 For example, consider the following string to be embedded in
243 the @ref{drawtext} filter description @option{text} value:
245 this is a 'string': may contain one, or more, special characters
248 This string contains the @code{'} special escaping character, and the
249 @code{:} special character, so it needs to be escaped in this way:
251 text=this is a \'string\'\: may contain one, or more, special characters
254 A second level of escaping is required when embedding the filter
255 description in a filtergraph description, in order to escape all the
256 filtergraph special characters. Thus the example above becomes:
258 drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
260 (note that in addition to the @code{\'} escaping special characters,
261 also @code{,} needs to be escaped).
263 Finally an additional level of escaping is needed when writing the
264 filtergraph description in a shell command, which depends on the
265 escaping rules of the adopted shell. For example, assuming that
266 @code{\} is special and needs to be escaped with another @code{\}, the
267 previous string will finally result in:
269 -vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
272 @chapter Timeline editing
274 Some filters support a generic @option{enable} option. For the filters
275 supporting timeline editing, this option can be set to an expression which is
276 evaluated before sending a frame to the filter. If the evaluation is non-zero,
277 the filter will be enabled, otherwise the frame will be sent unchanged to the
278 next filter in the filtergraph.
280 The expression accepts the following values:
283 timestamp expressed in seconds, NAN if the input timestamp is unknown
286 sequential number of the input frame, starting from 0
289 the position in the file of the input frame, NAN if unknown
293 width and height of the input frame if video
296 Additionally, these filters support an @option{enable} command that can be used
297 to re-define the expression.
299 Like any other filtering option, the @option{enable} option follows the same
302 For example, to enable a blur filter (@ref{smartblur}) from 10 seconds to 3
303 minutes, and a @ref{curves} filter starting at 3 seconds:
305 smartblur = enable='between(t,10,3*60)',
306 curves = enable='gte(t,3)' : preset=cross_process
309 See @code{ffmpeg -filters} to view which filters have timeline support.
311 @c man end FILTERGRAPH DESCRIPTION
313 @chapter Audio Filters
314 @c man begin AUDIO FILTERS
316 When you configure your FFmpeg build, you can disable any of the
317 existing filters using @code{--disable-filters}.
318 The configure output will show the audio filters included in your
321 Below is a description of the currently available audio filters.
325 A compressor is mainly used to reduce the dynamic range of a signal.
326 Especially modern music is mostly compressed at a high ratio to
327 improve the overall loudness. It's done to get the highest attention
328 of a listener, "fatten" the sound and bring more "power" to the track.
329 If a signal is compressed too much it may sound dull or "dead"
330 afterwards or it may start to "pump" (which could be a powerful effect
331 but can also destroy a track completely).
332 The right compression is the key to reach a professional sound and is
333 the high art of mixing and mastering. Because of its complex settings
334 it may take a long time to get the right feeling for this kind of effect.
336 Compression is done by detecting the volume above a chosen level
337 @code{threshold} and dividing it by the factor set with @code{ratio}.
338 So if you set the threshold to -12dB and your signal reaches -6dB a ratio
339 of 2:1 will result in a signal at -9dB. Because an exact manipulation of
340 the signal would cause distortion of the waveform the reduction can be
341 levelled over the time. This is done by setting "Attack" and "Release".
342 @code{attack} determines how long the signal has to rise above the threshold
343 before any reduction will occur and @code{release} sets the time the signal
344 has to fall below the threshold to reduce the reduction again. Shorter signals
345 than the chosen attack time will be left untouched.
346 The overall reduction of the signal can be made up afterwards with the
347 @code{makeup} setting. So compressing the peaks of a signal about 6dB and
348 raising the makeup to this level results in a signal twice as loud than the
349 source. To gain a softer entry in the compression the @code{knee} flattens the
350 hard edge at the threshold in the range of the chosen decibels.
352 The filter accepts the following options:
356 Set input gain. Default is 1. Range is between 0.015625 and 64.
359 If a signal of second stream rises above this level it will affect the gain
360 reduction of the first stream.
361 By default it is 0.125. Range is between 0.00097563 and 1.
364 Set a ratio by which the signal is reduced. 1:2 means that if the level
365 rose 4dB above the threshold, it will be only 2dB above after the reduction.
366 Default is 2. Range is between 1 and 20.
369 Amount of milliseconds the signal has to rise above the threshold before gain
370 reduction starts. Default is 20. Range is between 0.01 and 2000.
373 Amount of milliseconds the signal has to fall below the threshold before
374 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
377 Set the amount by how much signal will be amplified after processing.
378 Default is 2. Range is from 1 and 64.
381 Curve the sharp knee around the threshold to enter gain reduction more softly.
382 Default is 2.82843. Range is between 1 and 8.
385 Choose if the @code{average} level between all channels of input stream
386 or the louder(@code{maximum}) channel of input stream affects the
387 reduction. Default is @code{average}.
390 Should the exact signal be taken in case of @code{peak} or an RMS one in case
391 of @code{rms}. Default is @code{rms} which is mostly smoother.
394 How much to use compressed signal in output. Default is 1.
395 Range is between 0 and 1.
400 Apply cross fade from one input audio stream to another input audio stream.
401 The cross fade is applied for specified duration near the end of first stream.
403 The filter accepts the following options:
407 Specify the number of samples for which the cross fade effect has to last.
408 At the end of the cross fade effect the first input audio will be completely
409 silent. Default is 44100.
412 Specify the duration of the cross fade effect. See
413 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
414 for the accepted syntax.
415 By default the duration is determined by @var{nb_samples}.
416 If set this option is used instead of @var{nb_samples}.
419 Should first stream end overlap with second stream start. Default is enabled.
422 Set curve for cross fade transition for first stream.
425 Set curve for cross fade transition for second stream.
427 For description of available curve types see @ref{afade} filter description.
434 Cross fade from one input to another:
436 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:c1=exp:c2=exp output.flac
440 Cross fade from one input to another but without overlapping:
442 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:o=0:c1=exp:c2=exp output.flac
448 Reduce audio bit resolution.
450 This filter is bit crusher with enhanced functionality. A bit crusher
451 is used to audibly reduce number of bits an audio signal is sampled
452 with. This doesn't change the bit depth at all, it just produces the
453 effect. Material reduced in bit depth sounds more harsh and "digital".
454 This filter is able to even round to continuous values instead of discrete
456 Additionally it has a D/C offset which results in different crushing of
457 the lower and the upper half of the signal.
458 An Anti-Aliasing setting is able to produce "softer" crushing sounds.
460 Another feature of this filter is the logarithmic mode.
461 This setting switches from linear distances between bits to logarithmic ones.
462 The result is a much more "natural" sounding crusher which doesn't gate low
463 signals for example. The human ear has a logarithmic perception, too
464 so this kind of crushing is much more pleasant.
465 Logarithmic crushing is also able to get anti-aliased.
467 The filter accepts the following options:
483 Can be linear: @code{lin} or logarithmic: @code{log}.
492 Set sample reduction.
495 Enable LFO. By default disabled.
506 Delay one or more audio channels.
508 Samples in delayed channel are filled with silence.
510 The filter accepts the following option:
514 Set list of delays in milliseconds for each channel separated by '|'.
515 At least one delay greater than 0 should be provided.
516 Unused delays will be silently ignored. If number of given delays is
517 smaller than number of channels all remaining channels will not be delayed.
518 If you want to delay exact number of samples, append 'S' to number.
525 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
526 the second channel (and any other channels that may be present) unchanged.
532 Delay second channel by 500 samples, the third channel by 700 samples and leave
533 the first channel (and any other channels that may be present) unchanged.
541 Apply echoing to the input audio.
543 Echoes are reflected sound and can occur naturally amongst mountains
544 (and sometimes large buildings) when talking or shouting; digital echo
545 effects emulate this behaviour and are often used to help fill out the
546 sound of a single instrument or vocal. The time difference between the
547 original signal and the reflection is the @code{delay}, and the
548 loudness of the reflected signal is the @code{decay}.
549 Multiple echoes can have different delays and decays.
551 A description of the accepted parameters follows.
555 Set input gain of reflected signal. Default is @code{0.6}.
558 Set output gain of reflected signal. Default is @code{0.3}.
561 Set list of time intervals in milliseconds between original signal and reflections
562 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
563 Default is @code{1000}.
566 Set list of loudnesses of reflected signals separated by '|'.
567 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
568 Default is @code{0.5}.
575 Make it sound as if there are twice as many instruments as are actually playing:
577 aecho=0.8:0.88:60:0.4
581 If delay is very short, then it sound like a (metallic) robot playing music:
587 A longer delay will sound like an open air concert in the mountains:
589 aecho=0.8:0.9:1000:0.3
593 Same as above but with one more mountain:
595 aecho=0.8:0.9:1000|1800:0.3|0.25
600 Audio emphasis filter creates or restores material directly taken from LPs or
601 emphased CDs with different filter curves. E.g. to store music on vinyl the
602 signal has to be altered by a filter first to even out the disadvantages of
603 this recording medium.
604 Once the material is played back the inverse filter has to be applied to
605 restore the distortion of the frequency response.
607 The filter accepts the following options:
617 Set filter mode. For restoring material use @code{reproduction} mode, otherwise
618 use @code{production} mode. Default is @code{reproduction} mode.
621 Set filter type. Selects medium. Can be one of the following:
633 select Compact Disc (CD).
639 select 50µs (FM-KF).
641 select 75µs (FM-KF).
647 Modify an audio signal according to the specified expressions.
649 This filter accepts one or more expressions (one for each channel),
650 which are evaluated and used to modify a corresponding audio signal.
652 It accepts the following parameters:
656 Set the '|'-separated expressions list for each separate channel. If
657 the number of input channels is greater than the number of
658 expressions, the last specified expression is used for the remaining
661 @item channel_layout, c
662 Set output channel layout. If not specified, the channel layout is
663 specified by the number of expressions. If set to @samp{same}, it will
664 use by default the same input channel layout.
667 Each expression in @var{exprs} can contain the following constants and functions:
671 channel number of the current expression
674 number of the evaluated sample, starting from 0
680 time of the evaluated sample expressed in seconds
683 @item nb_out_channels
684 input and output number of channels
687 the value of input channel with number @var{CH}
690 Note: this filter is slow. For faster processing you should use a
699 aeval=val(ch)/2:c=same
703 Invert phase of the second channel:
712 Apply fade-in/out effect to input audio.
714 A description of the accepted parameters follows.
718 Specify the effect type, can be either @code{in} for fade-in, or
719 @code{out} for a fade-out effect. Default is @code{in}.
721 @item start_sample, ss
722 Specify the number of the start sample for starting to apply the fade
723 effect. Default is 0.
726 Specify the number of samples for which the fade effect has to last. At
727 the end of the fade-in effect the output audio will have the same
728 volume as the input audio, at the end of the fade-out transition
729 the output audio will be silence. Default is 44100.
732 Specify the start time of the fade effect. Default is 0.
733 The value must be specified as a time duration; see
734 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
735 for the accepted syntax.
736 If set this option is used instead of @var{start_sample}.
739 Specify the duration of the fade effect. See
740 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
741 for the accepted syntax.
742 At the end of the fade-in effect the output audio will have the same
743 volume as the input audio, at the end of the fade-out transition
744 the output audio will be silence.
745 By default the duration is determined by @var{nb_samples}.
746 If set this option is used instead of @var{nb_samples}.
749 Set curve for fade transition.
751 It accepts the following values:
754 select triangular, linear slope (default)
756 select quarter of sine wave
758 select half of sine wave
760 select exponential sine wave
764 select inverted parabola
778 select inverted quarter of sine wave
780 select inverted half of sine wave
782 select double-exponential seat
784 select double-exponential sigmoid
792 Fade in first 15 seconds of audio:
798 Fade out last 25 seconds of a 900 seconds audio:
800 afade=t=out:st=875:d=25
805 Apply arbitrary expressions to samples in frequency domain.
809 Set frequency domain real expression for each separate channel separated
810 by '|'. Default is "1".
811 If the number of input channels is greater than the number of
812 expressions, the last specified expression is used for the remaining
816 Set frequency domain imaginary expression for each separate channel
817 separated by '|'. If not set, @var{real} option is used.
819 Each expression in @var{real} and @var{imag} can contain the following
827 current frequency bin number
830 number of available bins
833 channel number of the current expression
845 It accepts the following values:
861 Default is @code{w4096}
864 Set window function. Default is @code{hann}.
867 Set window overlap. If set to 1, the recommended overlap for selected
868 window function will be picked. Default is @code{0.75}.
875 Leave almost only low frequencies in audio:
877 afftfilt="1-clip((b/nb)*b,0,1)"
883 Apply an arbitrary Frequency Impulse Response filter.
885 This filter is designed for applying long FIR filters,
886 up to 30 seconds long.
888 It can be used as component for digital crossover filters,
889 room equalization, cross talk cancellation, wavefield synthesis,
890 auralization, ambiophonics and ambisonics.
892 This filter uses second stream as FIR coefficients.
893 If second stream holds single channel, it will be used
894 for all input channels in first stream, otherwise
895 number of channels in second stream must be same as
896 number of channels in first stream.
898 It accepts the following parameters:
902 Set dry gain. This sets input gain.
905 Set wet gain. This sets final output gain.
908 Set Impulse Response filter length. Default is 1, which means whole IR is processed.
911 Enable applying gain measured from power of IR.
918 Apply reverb to stream using mono IR file as second input, complete command using ffmpeg:
920 ffmpeg -i input.wav -i middle_tunnel_1way_mono.wav -lavfi afir output.wav
927 Set output format constraints for the input audio. The framework will
928 negotiate the most appropriate format to minimize conversions.
930 It accepts the following parameters:
934 A '|'-separated list of requested sample formats.
937 A '|'-separated list of requested sample rates.
939 @item channel_layouts
940 A '|'-separated list of requested channel layouts.
942 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
943 for the required syntax.
946 If a parameter is omitted, all values are allowed.
948 Force the output to either unsigned 8-bit or signed 16-bit stereo
950 aformat=sample_fmts=u8|s16:channel_layouts=stereo
955 A gate is mainly used to reduce lower parts of a signal. This kind of signal
956 processing reduces disturbing noise between useful signals.
958 Gating is done by detecting the volume below a chosen level @var{threshold}
959 and dividing it by the factor set with @var{ratio}. The bottom of the noise
960 floor is set via @var{range}. Because an exact manipulation of the signal
961 would cause distortion of the waveform the reduction can be levelled over
962 time. This is done by setting @var{attack} and @var{release}.
964 @var{attack} determines how long the signal has to fall below the threshold
965 before any reduction will occur and @var{release} sets the time the signal
966 has to rise above the threshold to reduce the reduction again.
967 Shorter signals than the chosen attack time will be left untouched.
971 Set input level before filtering.
972 Default is 1. Allowed range is from 0.015625 to 64.
975 Set the level of gain reduction when the signal is below the threshold.
976 Default is 0.06125. Allowed range is from 0 to 1.
979 If a signal rises above this level the gain reduction is released.
980 Default is 0.125. Allowed range is from 0 to 1.
983 Set a ratio by which the signal is reduced.
984 Default is 2. Allowed range is from 1 to 9000.
987 Amount of milliseconds the signal has to rise above the threshold before gain
989 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
992 Amount of milliseconds the signal has to fall below the threshold before the
993 reduction is increased again. Default is 250 milliseconds.
994 Allowed range is from 0.01 to 9000.
997 Set amount of amplification of signal after processing.
998 Default is 1. Allowed range is from 1 to 64.
1001 Curve the sharp knee around the threshold to enter gain reduction more softly.
1002 Default is 2.828427125. Allowed range is from 1 to 8.
1005 Choose if exact signal should be taken for detection or an RMS like one.
1006 Default is @code{rms}. Can be @code{peak} or @code{rms}.
1009 Choose if the average level between all channels or the louder channel affects
1011 Default is @code{average}. Can be @code{average} or @code{maximum}.
1016 The limiter prevents an input signal from rising over a desired threshold.
1017 This limiter uses lookahead technology to prevent your signal from distorting.
1018 It means that there is a small delay after the signal is processed. Keep in mind
1019 that the delay it produces is the attack time you set.
1021 The filter accepts the following options:
1025 Set input gain. Default is 1.
1028 Set output gain. Default is 1.
1031 Don't let signals above this level pass the limiter. Default is 1.
1034 The limiter will reach its attenuation level in this amount of time in
1035 milliseconds. Default is 5 milliseconds.
1038 Come back from limiting to attenuation 1.0 in this amount of milliseconds.
1039 Default is 50 milliseconds.
1042 When gain reduction is always needed ASC takes care of releasing to an
1043 average reduction level rather than reaching a reduction of 0 in the release
1047 Select how much the release time is affected by ASC, 0 means nearly no changes
1048 in release time while 1 produces higher release times.
1051 Auto level output signal. Default is enabled.
1052 This normalizes audio back to 0dB if enabled.
1055 Depending on picked setting it is recommended to upsample input 2x or 4x times
1056 with @ref{aresample} before applying this filter.
1060 Apply a two-pole all-pass filter with central frequency (in Hz)
1061 @var{frequency}, and filter-width @var{width}.
1062 An all-pass filter changes the audio's frequency to phase relationship
1063 without changing its frequency to amplitude relationship.
1065 The filter accepts the following options:
1069 Set frequency in Hz.
1072 Set method to specify band-width of filter.
1085 Specify the band-width of a filter in width_type units.
1088 Specify which channels to filter, by default all available are filtered.
1095 The filter accepts the following options:
1099 Set the number of loops.
1102 Set maximal number of samples.
1105 Set first sample of loop.
1111 Merge two or more audio streams into a single multi-channel stream.
1113 The filter accepts the following options:
1118 Set the number of inputs. Default is 2.
1122 If the channel layouts of the inputs are disjoint, and therefore compatible,
1123 the channel layout of the output will be set accordingly and the channels
1124 will be reordered as necessary. If the channel layouts of the inputs are not
1125 disjoint, the output will have all the channels of the first input then all
1126 the channels of the second input, in that order, and the channel layout of
1127 the output will be the default value corresponding to the total number of
1130 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
1131 is FC+BL+BR, then the output will be in 5.1, with the channels in the
1132 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
1133 first input, b1 is the first channel of the second input).
1135 On the other hand, if both input are in stereo, the output channels will be
1136 in the default order: a1, a2, b1, b2, and the channel layout will be
1137 arbitrarily set to 4.0, which may or may not be the expected value.
1139 All inputs must have the same sample rate, and format.
1141 If inputs do not have the same duration, the output will stop with the
1144 @subsection Examples
1148 Merge two mono files into a stereo stream:
1150 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
1154 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
1156 ffmpeg -i input.mkv -filter_complex "[0:1][0:2][0:3][0:4][0:5][0:6] amerge=inputs=6" -c:a pcm_s16le output.mkv
1162 Mixes multiple audio inputs into a single output.
1164 Note that this filter only supports float samples (the @var{amerge}
1165 and @var{pan} audio filters support many formats). If the @var{amix}
1166 input has integer samples then @ref{aresample} will be automatically
1167 inserted to perform the conversion to float samples.
1171 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
1173 will mix 3 input audio streams to a single output with the same duration as the
1174 first input and a dropout transition time of 3 seconds.
1176 It accepts the following parameters:
1180 The number of inputs. If unspecified, it defaults to 2.
1183 How to determine the end-of-stream.
1187 The duration of the longest input. (default)
1190 The duration of the shortest input.
1193 The duration of the first input.
1197 @item dropout_transition
1198 The transition time, in seconds, for volume renormalization when an input
1199 stream ends. The default value is 2 seconds.
1203 @section anequalizer
1205 High-order parametric multiband equalizer for each channel.
1207 It accepts the following parameters:
1211 This option string is in format:
1212 "c@var{chn} f=@var{cf} w=@var{w} g=@var{g} t=@var{f} | ..."
1213 Each equalizer band is separated by '|'.
1217 Set channel number to which equalization will be applied.
1218 If input doesn't have that channel the entry is ignored.
1221 Set central frequency for band.
1222 If input doesn't have that frequency the entry is ignored.
1225 Set band width in hertz.
1228 Set band gain in dB.
1231 Set filter type for band, optional, can be:
1235 Butterworth, this is default.
1246 With this option activated frequency response of anequalizer is displayed
1250 Set video stream size. Only useful if curves option is activated.
1253 Set max gain that will be displayed. Only useful if curves option is activated.
1254 Setting this to a reasonable value makes it possible to display gain which is derived from
1255 neighbour bands which are too close to each other and thus produce higher gain
1256 when both are activated.
1259 Set frequency scale used to draw frequency response in video output.
1260 Can be linear or logarithmic. Default is logarithmic.
1263 Set color for each channel curve which is going to be displayed in video stream.
1264 This is list of color names separated by space or by '|'.
1265 Unrecognised or missing colors will be replaced by white color.
1268 @subsection Examples
1272 Lower gain by 10 of central frequency 200Hz and width 100 Hz
1273 for first 2 channels using Chebyshev type 1 filter:
1275 anequalizer=c0 f=200 w=100 g=-10 t=1|c1 f=200 w=100 g=-10 t=1
1279 @subsection Commands
1281 This filter supports the following commands:
1284 Alter existing filter parameters.
1285 Syntax for the commands is : "@var{fN}|f=@var{freq}|w=@var{width}|g=@var{gain}"
1287 @var{fN} is existing filter number, starting from 0, if no such filter is available
1289 @var{freq} set new frequency parameter.
1290 @var{width} set new width parameter in herz.
1291 @var{gain} set new gain parameter in dB.
1293 Full filter invocation with asendcmd may look like this:
1294 asendcmd=c='4.0 anequalizer change 0|f=200|w=50|g=1',anequalizer=...
1299 Pass the audio source unchanged to the output.
1303 Pad the end of an audio stream with silence.
1305 This can be used together with @command{ffmpeg} @option{-shortest} to
1306 extend audio streams to the same length as the video stream.
1308 A description of the accepted options follows.
1312 Set silence packet size. Default value is 4096.
1315 Set the number of samples of silence to add to the end. After the
1316 value is reached, the stream is terminated. This option is mutually
1317 exclusive with @option{whole_len}.
1320 Set the minimum total number of samples in the output audio stream. If
1321 the value is longer than the input audio length, silence is added to
1322 the end, until the value is reached. This option is mutually exclusive
1323 with @option{pad_len}.
1326 If neither the @option{pad_len} nor the @option{whole_len} option is
1327 set, the filter will add silence to the end of the input stream
1330 @subsection Examples
1334 Add 1024 samples of silence to the end of the input:
1340 Make sure the audio output will contain at least 10000 samples, pad
1341 the input with silence if required:
1343 apad=whole_len=10000
1347 Use @command{ffmpeg} to pad the audio input with silence, so that the
1348 video stream will always result the shortest and will be converted
1349 until the end in the output file when using the @option{shortest}
1352 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
1357 Add a phasing effect to the input audio.
1359 A phaser filter creates series of peaks and troughs in the frequency spectrum.
1360 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
1362 A description of the accepted parameters follows.
1366 Set input gain. Default is 0.4.
1369 Set output gain. Default is 0.74
1372 Set delay in milliseconds. Default is 3.0.
1375 Set decay. Default is 0.4.
1378 Set modulation speed in Hz. Default is 0.5.
1381 Set modulation type. Default is triangular.
1383 It accepts the following values:
1392 Audio pulsator is something between an autopanner and a tremolo.
1393 But it can produce funny stereo effects as well. Pulsator changes the volume
1394 of the left and right channel based on a LFO (low frequency oscillator) with
1395 different waveforms and shifted phases.
1396 This filter have the ability to define an offset between left and right
1397 channel. An offset of 0 means that both LFO shapes match each other.
1398 The left and right channel are altered equally - a conventional tremolo.
1399 An offset of 50% means that the shape of the right channel is exactly shifted
1400 in phase (or moved backwards about half of the frequency) - pulsator acts as
1401 an autopanner. At 1 both curves match again. Every setting in between moves the
1402 phase shift gapless between all stages and produces some "bypassing" sounds with
1403 sine and triangle waveforms. The more you set the offset near 1 (starting from
1404 the 0.5) the faster the signal passes from the left to the right speaker.
1406 The filter accepts the following options:
1410 Set input gain. By default it is 1. Range is [0.015625 - 64].
1413 Set output gain. By default it is 1. Range is [0.015625 - 64].
1416 Set waveform shape the LFO will use. Can be one of: sine, triangle, square,
1417 sawup or sawdown. Default is sine.
1420 Set modulation. Define how much of original signal is affected by the LFO.
1423 Set left channel offset. Default is 0. Allowed range is [0 - 1].
1426 Set right channel offset. Default is 0.5. Allowed range is [0 - 1].
1429 Set pulse width. Default is 1. Allowed range is [0 - 2].
1432 Set possible timing mode. Can be one of: bpm, ms or hz. Default is hz.
1435 Set bpm. Default is 120. Allowed range is [30 - 300]. Only used if timing
1439 Set ms. Default is 500. Allowed range is [10 - 2000]. Only used if timing
1443 Set frequency in Hz. Default is 2. Allowed range is [0.01 - 100]. Only used
1444 if timing is set to hz.
1450 Resample the input audio to the specified parameters, using the
1451 libswresample library. If none are specified then the filter will
1452 automatically convert between its input and output.
1454 This filter is also able to stretch/squeeze the audio data to make it match
1455 the timestamps or to inject silence / cut out audio to make it match the
1456 timestamps, do a combination of both or do neither.
1458 The filter accepts the syntax
1459 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
1460 expresses a sample rate and @var{resampler_options} is a list of
1461 @var{key}=@var{value} pairs, separated by ":". See the
1462 @ref{Resampler Options,,the "Resampler Options" section in the
1463 ffmpeg-resampler(1) manual,ffmpeg-resampler}
1464 for the complete list of supported options.
1466 @subsection Examples
1470 Resample the input audio to 44100Hz:
1476 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
1477 samples per second compensation:
1479 aresample=async=1000
1485 Reverse an audio clip.
1487 Warning: This filter requires memory to buffer the entire clip, so trimming
1490 @subsection Examples
1494 Take the first 5 seconds of a clip, and reverse it.
1496 atrim=end=5,areverse
1500 @section asetnsamples
1502 Set the number of samples per each output audio frame.
1504 The last output packet may contain a different number of samples, as
1505 the filter will flush all the remaining samples when the input audio
1508 The filter accepts the following options:
1512 @item nb_out_samples, n
1513 Set the number of frames per each output audio frame. The number is
1514 intended as the number of samples @emph{per each channel}.
1515 Default value is 1024.
1518 If set to 1, the filter will pad the last audio frame with zeroes, so
1519 that the last frame will contain the same number of samples as the
1520 previous ones. Default value is 1.
1523 For example, to set the number of per-frame samples to 1234 and
1524 disable padding for the last frame, use:
1526 asetnsamples=n=1234:p=0
1531 Set the sample rate without altering the PCM data.
1532 This will result in a change of speed and pitch.
1534 The filter accepts the following options:
1537 @item sample_rate, r
1538 Set the output sample rate. Default is 44100 Hz.
1543 Show a line containing various information for each input audio frame.
1544 The input audio is not modified.
1546 The shown line contains a sequence of key/value pairs of the form
1547 @var{key}:@var{value}.
1549 The following values are shown in the output:
1553 The (sequential) number of the input frame, starting from 0.
1556 The presentation timestamp of the input frame, in time base units; the time base
1557 depends on the filter input pad, and is usually 1/@var{sample_rate}.
1560 The presentation timestamp of the input frame in seconds.
1563 position of the frame in the input stream, -1 if this information in
1564 unavailable and/or meaningless (for example in case of synthetic audio)
1573 The sample rate for the audio frame.
1576 The number of samples (per channel) in the frame.
1579 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
1580 audio, the data is treated as if all the planes were concatenated.
1582 @item plane_checksums
1583 A list of Adler-32 checksums for each data plane.
1589 Display time domain statistical information about the audio channels.
1590 Statistics are calculated and displayed for each audio channel and,
1591 where applicable, an overall figure is also given.
1593 It accepts the following option:
1596 Short window length in seconds, used for peak and trough RMS measurement.
1597 Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
1601 Set metadata injection. All the metadata keys are prefixed with @code{lavfi.astats.X},
1602 where @code{X} is channel number starting from 1 or string @code{Overall}. Default is
1605 Available keys for each channel are:
1636 For example full key look like this @code{lavfi.astats.1.DC_offset} or
1637 this @code{lavfi.astats.Overall.Peak_count}.
1639 For description what each key means read below.
1642 Set number of frame after which stats are going to be recalculated.
1643 Default is disabled.
1646 A description of each shown parameter follows:
1650 Mean amplitude displacement from zero.
1653 Minimal sample level.
1656 Maximal sample level.
1658 @item Min difference
1659 Minimal difference between two consecutive samples.
1661 @item Max difference
1662 Maximal difference between two consecutive samples.
1664 @item Mean difference
1665 Mean difference between two consecutive samples.
1666 The average of each difference between two consecutive samples.
1670 Standard peak and RMS level measured in dBFS.
1674 Peak and trough values for RMS level measured over a short window.
1677 Standard ratio of peak to RMS level (note: not in dB).
1680 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
1681 (i.e. either @var{Min level} or @var{Max level}).
1684 Number of occasions (not the number of samples) that the signal attained either
1685 @var{Min level} or @var{Max level}.
1688 Overall bit depth of audio. Number of bits used for each sample.
1695 The filter accepts exactly one parameter, the audio tempo. If not
1696 specified then the filter will assume nominal 1.0 tempo. Tempo must
1697 be in the [0.5, 2.0] range.
1699 @subsection Examples
1703 Slow down audio to 80% tempo:
1709 To speed up audio to 125% tempo:
1717 Trim the input so that the output contains one continuous subpart of the input.
1719 It accepts the following parameters:
1722 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1723 sample with the timestamp @var{start} will be the first sample in the output.
1726 Specify time of the first audio sample that will be dropped, i.e. the
1727 audio sample immediately preceding the one with the timestamp @var{end} will be
1728 the last sample in the output.
1731 Same as @var{start}, except this option sets the start timestamp in samples
1735 Same as @var{end}, except this option sets the end timestamp in samples instead
1739 The maximum duration of the output in seconds.
1742 The number of the first sample that should be output.
1745 The number of the first sample that should be dropped.
1748 @option{start}, @option{end}, and @option{duration} are expressed as time
1749 duration specifications; see
1750 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1752 Note that the first two sets of the start/end options and the @option{duration}
1753 option look at the frame timestamp, while the _sample options simply count the
1754 samples that pass through the filter. So start/end_pts and start/end_sample will
1755 give different results when the timestamps are wrong, inexact or do not start at
1756 zero. Also note that this filter does not modify the timestamps. If you wish
1757 to have the output timestamps start at zero, insert the asetpts filter after the
1760 If multiple start or end options are set, this filter tries to be greedy and
1761 keep all samples that match at least one of the specified constraints. To keep
1762 only the part that matches all the constraints at once, chain multiple atrim
1765 The defaults are such that all the input is kept. So it is possible to set e.g.
1766 just the end values to keep everything before the specified time.
1771 Drop everything except the second minute of input:
1773 ffmpeg -i INPUT -af atrim=60:120
1777 Keep only the first 1000 samples:
1779 ffmpeg -i INPUT -af atrim=end_sample=1000
1786 Apply a two-pole Butterworth band-pass filter with central
1787 frequency @var{frequency}, and (3dB-point) band-width width.
1788 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1789 instead of the default: constant 0dB peak gain.
1790 The filter roll off at 6dB per octave (20dB per decade).
1792 The filter accepts the following options:
1796 Set the filter's central frequency. Default is @code{3000}.
1799 Constant skirt gain if set to 1. Defaults to 0.
1802 Set method to specify band-width of filter.
1815 Specify the band-width of a filter in width_type units.
1818 Specify which channels to filter, by default all available are filtered.
1823 Apply a two-pole Butterworth band-reject filter with central
1824 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1825 The filter roll off at 6dB per octave (20dB per decade).
1827 The filter accepts the following options:
1831 Set the filter's central frequency. Default is @code{3000}.
1834 Set method to specify band-width of filter.
1847 Specify the band-width of a filter in width_type units.
1850 Specify which channels to filter, by default all available are filtered.
1855 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1856 shelving filter with a response similar to that of a standard
1857 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1859 The filter accepts the following options:
1863 Give the gain at 0 Hz. Its useful range is about -20
1864 (for a large cut) to +20 (for a large boost).
1865 Beware of clipping when using a positive gain.
1868 Set the filter's central frequency and so can be used
1869 to extend or reduce the frequency range to be boosted or cut.
1870 The default value is @code{100} Hz.
1873 Set method to specify band-width of filter.
1886 Determine how steep is the filter's shelf transition.
1889 Specify which channels to filter, by default all available are filtered.
1894 Apply a biquad IIR filter with the given coefficients.
1895 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1896 are the numerator and denominator coefficients respectively.
1897 and @var{channels}, @var{c} specify which channels to filter, by default all
1898 available are filtered.
1901 Bauer stereo to binaural transformation, which improves headphone listening of
1902 stereo audio records.
1904 It accepts the following parameters:
1908 Pre-defined crossfeed level.
1912 Default level (fcut=700, feed=50).
1915 Chu Moy circuit (fcut=700, feed=60).
1918 Jan Meier circuit (fcut=650, feed=95).
1923 Cut frequency (in Hz).
1932 Remap input channels to new locations.
1934 It accepts the following parameters:
1937 Map channels from input to output. The argument is a '|'-separated list of
1938 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1939 @var{in_channel} form. @var{in_channel} can be either the name of the input
1940 channel (e.g. FL for front left) or its index in the input channel layout.
1941 @var{out_channel} is the name of the output channel or its index in the output
1942 channel layout. If @var{out_channel} is not given then it is implicitly an
1943 index, starting with zero and increasing by one for each mapping.
1945 @item channel_layout
1946 The channel layout of the output stream.
1949 If no mapping is present, the filter will implicitly map input channels to
1950 output channels, preserving indices.
1952 For example, assuming a 5.1+downmix input MOV file,
1954 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1956 will create an output WAV file tagged as stereo from the downmix channels of
1959 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1961 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1964 @section channelsplit
1966 Split each channel from an input audio stream into a separate output stream.
1968 It accepts the following parameters:
1970 @item channel_layout
1971 The channel layout of the input stream. The default is "stereo".
1974 For example, assuming a stereo input MP3 file,
1976 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1978 will create an output Matroska file with two audio streams, one containing only
1979 the left channel and the other the right channel.
1981 Split a 5.1 WAV file into per-channel files:
1983 ffmpeg -i in.wav -filter_complex
1984 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1985 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1986 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1991 Add a chorus effect to the audio.
1993 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1995 Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1996 constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1997 The modulation depth defines the range the modulated delay is played before or after
1998 the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1999 sound tuned around the original one, like in a chorus where some vocals are slightly
2002 It accepts the following parameters:
2005 Set input gain. Default is 0.4.
2008 Set output gain. Default is 0.4.
2011 Set delays. A typical delay is around 40ms to 60ms.
2023 @subsection Examples
2029 chorus=0.7:0.9:55:0.4:0.25:2
2035 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
2039 Fuller sounding chorus with three delays:
2041 chorus=0.5:0.9:50|60|40:0.4|0.32|0.3:0.25|0.4|0.3:2|2.3|1.3
2046 Compress or expand the audio's dynamic range.
2048 It accepts the following parameters:
2054 A list of times in seconds for each channel over which the instantaneous level
2055 of the input signal is averaged to determine its volume. @var{attacks} refers to
2056 increase of volume and @var{decays} refers to decrease of volume. For most
2057 situations, the attack time (response to the audio getting louder) should be
2058 shorter than the decay time, because the human ear is more sensitive to sudden
2059 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
2060 a typical value for decay is 0.8 seconds.
2061 If specified number of attacks & decays is lower than number of channels, the last
2062 set attack/decay will be used for all remaining channels.
2065 A list of points for the transfer function, specified in dB relative to the
2066 maximum possible signal amplitude. Each key points list must be defined using
2067 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
2068 @code{x0/y0 x1/y1 x2/y2 ....}
2070 The input values must be in strictly increasing order but the transfer function
2071 does not have to be monotonically rising. The point @code{0/0} is assumed but
2072 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
2073 function are @code{-70/-70|-60/-20}.
2076 Set the curve radius in dB for all joints. It defaults to 0.01.
2079 Set the additional gain in dB to be applied at all points on the transfer
2080 function. This allows for easy adjustment of the overall gain.
2084 Set an initial volume, in dB, to be assumed for each channel when filtering
2085 starts. This permits the user to supply a nominal level initially, so that, for
2086 example, a very large gain is not applied to initial signal levels before the
2087 companding has begun to operate. A typical value for audio which is initially
2088 quiet is -90 dB. It defaults to 0.
2091 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
2092 delayed before being fed to the volume adjuster. Specifying a delay
2093 approximately equal to the attack/decay times allows the filter to effectively
2094 operate in predictive rather than reactive mode. It defaults to 0.
2098 @subsection Examples
2102 Make music with both quiet and loud passages suitable for listening to in a
2105 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
2108 Another example for audio with whisper and explosion parts:
2110 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
2114 A noise gate for when the noise is at a lower level than the signal:
2116 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
2120 Here is another noise gate, this time for when the noise is at a higher level
2121 than the signal (making it, in some ways, similar to squelch):
2123 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
2127 2:1 compression starting at -6dB:
2129 compand=points=-80/-80|-6/-6|0/-3.8|20/3.5
2133 2:1 compression starting at -9dB:
2135 compand=points=-80/-80|-9/-9|0/-5.3|20/2.9
2139 2:1 compression starting at -12dB:
2141 compand=points=-80/-80|-12/-12|0/-6.8|20/1.9
2145 2:1 compression starting at -18dB:
2147 compand=points=-80/-80|-18/-18|0/-9.8|20/0.7
2151 3:1 compression starting at -15dB:
2153 compand=points=-80/-80|-15/-15|0/-10.8|20/-5.2
2159 compand=points=-80/-105|-62/-80|-15.4/-15.4|0/-12|20/-7.6
2165 compand=attacks=0:points=-80/-169|-54/-80|-49.5/-64.6|-41.1/-41.1|-25.8/-15|-10.8/-4.5|0/0|20/8.3
2169 Hard limiter at -6dB:
2171 compand=attacks=0:points=-80/-80|-6/-6|20/-6
2175 Hard limiter at -12dB:
2177 compand=attacks=0:points=-80/-80|-12/-12|20/-12
2181 Hard noise gate at -35 dB:
2183 compand=attacks=0:points=-80/-115|-35.1/-80|-35/-35|20/20
2189 compand=attacks=0:points=-80/-80|-12.4/-12.4|-6/-8|0/-6.8|20/-2.8
2193 @section compensationdelay
2195 Compensation Delay Line is a metric based delay to compensate differing
2196 positions of microphones or speakers.
2198 For example, you have recorded guitar with two microphones placed in
2199 different location. Because the front of sound wave has fixed speed in
2200 normal conditions, the phasing of microphones can vary and depends on
2201 their location and interposition. The best sound mix can be achieved when
2202 these microphones are in phase (synchronized). Note that distance of
2203 ~30 cm between microphones makes one microphone to capture signal in
2204 antiphase to another microphone. That makes the final mix sounding moody.
2205 This filter helps to solve phasing problems by adding different delays
2206 to each microphone track and make them synchronized.
2208 The best result can be reached when you take one track as base and
2209 synchronize other tracks one by one with it.
2210 Remember that synchronization/delay tolerance depends on sample rate, too.
2211 Higher sample rates will give more tolerance.
2213 It accepts the following parameters:
2217 Set millimeters distance. This is compensation distance for fine tuning.
2221 Set cm distance. This is compensation distance for tightening distance setup.
2225 Set meters distance. This is compensation distance for hard distance setup.
2229 Set dry amount. Amount of unprocessed (dry) signal.
2233 Set wet amount. Amount of processed (wet) signal.
2237 Set temperature degree in Celsius. This is the temperature of the environment.
2241 @section crystalizer
2242 Simple algorithm to expand audio dynamic range.
2244 The filter accepts the following options:
2248 Sets the intensity of effect (default: 2.0). Must be in range between 0.0
2249 (unchanged sound) to 10.0 (maximum effect).
2252 Enable clipping. By default is enabled.
2256 Apply a DC shift to the audio.
2258 This can be useful to remove a DC offset (caused perhaps by a hardware problem
2259 in the recording chain) from the audio. The effect of a DC offset is reduced
2260 headroom and hence volume. The @ref{astats} filter can be used to determine if
2261 a signal has a DC offset.
2265 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
2269 Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
2270 used to prevent clipping.
2274 Dynamic Audio Normalizer.
2276 This filter applies a certain amount of gain to the input audio in order
2277 to bring its peak magnitude to a target level (e.g. 0 dBFS). However, in
2278 contrast to more "simple" normalization algorithms, the Dynamic Audio
2279 Normalizer *dynamically* re-adjusts the gain factor to the input audio.
2280 This allows for applying extra gain to the "quiet" sections of the audio
2281 while avoiding distortions or clipping the "loud" sections. In other words:
2282 The Dynamic Audio Normalizer will "even out" the volume of quiet and loud
2283 sections, in the sense that the volume of each section is brought to the
2284 same target level. Note, however, that the Dynamic Audio Normalizer achieves
2285 this goal *without* applying "dynamic range compressing". It will retain 100%
2286 of the dynamic range *within* each section of the audio file.
2290 Set the frame length in milliseconds. In range from 10 to 8000 milliseconds.
2291 Default is 500 milliseconds.
2292 The Dynamic Audio Normalizer processes the input audio in small chunks,
2293 referred to as frames. This is required, because a peak magnitude has no
2294 meaning for just a single sample value. Instead, we need to determine the
2295 peak magnitude for a contiguous sequence of sample values. While a "standard"
2296 normalizer would simply use the peak magnitude of the complete file, the
2297 Dynamic Audio Normalizer determines the peak magnitude individually for each
2298 frame. The length of a frame is specified in milliseconds. By default, the
2299 Dynamic Audio Normalizer uses a frame length of 500 milliseconds, which has
2300 been found to give good results with most files.
2301 Note that the exact frame length, in number of samples, will be determined
2302 automatically, based on the sampling rate of the individual input audio file.
2305 Set the Gaussian filter window size. In range from 3 to 301, must be odd
2306 number. Default is 31.
2307 Probably the most important parameter of the Dynamic Audio Normalizer is the
2308 @code{window size} of the Gaussian smoothing filter. The filter's window size
2309 is specified in frames, centered around the current frame. For the sake of
2310 simplicity, this must be an odd number. Consequently, the default value of 31
2311 takes into account the current frame, as well as the 15 preceding frames and
2312 the 15 subsequent frames. Using a larger window results in a stronger
2313 smoothing effect and thus in less gain variation, i.e. slower gain
2314 adaptation. Conversely, using a smaller window results in a weaker smoothing
2315 effect and thus in more gain variation, i.e. faster gain adaptation.
2316 In other words, the more you increase this value, the more the Dynamic Audio
2317 Normalizer will behave like a "traditional" normalization filter. On the
2318 contrary, the more you decrease this value, the more the Dynamic Audio
2319 Normalizer will behave like a dynamic range compressor.
2322 Set the target peak value. This specifies the highest permissible magnitude
2323 level for the normalized audio input. This filter will try to approach the
2324 target peak magnitude as closely as possible, but at the same time it also
2325 makes sure that the normalized signal will never exceed the peak magnitude.
2326 A frame's maximum local gain factor is imposed directly by the target peak
2327 magnitude. The default value is 0.95 and thus leaves a headroom of 5%*.
2328 It is not recommended to go above this value.
2331 Set the maximum gain factor. In range from 1.0 to 100.0. Default is 10.0.
2332 The Dynamic Audio Normalizer determines the maximum possible (local) gain
2333 factor for each input frame, i.e. the maximum gain factor that does not
2334 result in clipping or distortion. The maximum gain factor is determined by
2335 the frame's highest magnitude sample. However, the Dynamic Audio Normalizer
2336 additionally bounds the frame's maximum gain factor by a predetermined
2337 (global) maximum gain factor. This is done in order to avoid excessive gain
2338 factors in "silent" or almost silent frames. By default, the maximum gain
2339 factor is 10.0, For most inputs the default value should be sufficient and
2340 it usually is not recommended to increase this value. Though, for input
2341 with an extremely low overall volume level, it may be necessary to allow even
2342 higher gain factors. Note, however, that the Dynamic Audio Normalizer does
2343 not simply apply a "hard" threshold (i.e. cut off values above the threshold).
2344 Instead, a "sigmoid" threshold function will be applied. This way, the
2345 gain factors will smoothly approach the threshold value, but never exceed that
2349 Set the target RMS. In range from 0.0 to 1.0. Default is 0.0 - disabled.
2350 By default, the Dynamic Audio Normalizer performs "peak" normalization.
2351 This means that the maximum local gain factor for each frame is defined
2352 (only) by the frame's highest magnitude sample. This way, the samples can
2353 be amplified as much as possible without exceeding the maximum signal
2354 level, i.e. without clipping. Optionally, however, the Dynamic Audio
2355 Normalizer can also take into account the frame's root mean square,
2356 abbreviated RMS. In electrical engineering, the RMS is commonly used to
2357 determine the power of a time-varying signal. It is therefore considered
2358 that the RMS is a better approximation of the "perceived loudness" than
2359 just looking at the signal's peak magnitude. Consequently, by adjusting all
2360 frames to a constant RMS value, a uniform "perceived loudness" can be
2361 established. If a target RMS value has been specified, a frame's local gain
2362 factor is defined as the factor that would result in exactly that RMS value.
2363 Note, however, that the maximum local gain factor is still restricted by the
2364 frame's highest magnitude sample, in order to prevent clipping.
2367 Enable channels coupling. By default is enabled.
2368 By default, the Dynamic Audio Normalizer will amplify all channels by the same
2369 amount. This means the same gain factor will be applied to all channels, i.e.
2370 the maximum possible gain factor is determined by the "loudest" channel.
2371 However, in some recordings, it may happen that the volume of the different
2372 channels is uneven, e.g. one channel may be "quieter" than the other one(s).
2373 In this case, this option can be used to disable the channel coupling. This way,
2374 the gain factor will be determined independently for each channel, depending
2375 only on the individual channel's highest magnitude sample. This allows for
2376 harmonizing the volume of the different channels.
2379 Enable DC bias correction. By default is disabled.
2380 An audio signal (in the time domain) is a sequence of sample values.
2381 In the Dynamic Audio Normalizer these sample values are represented in the
2382 -1.0 to 1.0 range, regardless of the original input format. Normally, the
2383 audio signal, or "waveform", should be centered around the zero point.
2384 That means if we calculate the mean value of all samples in a file, or in a
2385 single frame, then the result should be 0.0 or at least very close to that
2386 value. If, however, there is a significant deviation of the mean value from
2387 0.0, in either positive or negative direction, this is referred to as a
2388 DC bias or DC offset. Since a DC bias is clearly undesirable, the Dynamic
2389 Audio Normalizer provides optional DC bias correction.
2390 With DC bias correction enabled, the Dynamic Audio Normalizer will determine
2391 the mean value, or "DC correction" offset, of each input frame and subtract
2392 that value from all of the frame's sample values which ensures those samples
2393 are centered around 0.0 again. Also, in order to avoid "gaps" at the frame
2394 boundaries, the DC correction offset values will be interpolated smoothly
2395 between neighbouring frames.
2398 Enable alternative boundary mode. By default is disabled.
2399 The Dynamic Audio Normalizer takes into account a certain neighbourhood
2400 around each frame. This includes the preceding frames as well as the
2401 subsequent frames. However, for the "boundary" frames, located at the very
2402 beginning and at the very end of the audio file, not all neighbouring
2403 frames are available. In particular, for the first few frames in the audio
2404 file, the preceding frames are not known. And, similarly, for the last few
2405 frames in the audio file, the subsequent frames are not known. Thus, the
2406 question arises which gain factors should be assumed for the missing frames
2407 in the "boundary" region. The Dynamic Audio Normalizer implements two modes
2408 to deal with this situation. The default boundary mode assumes a gain factor
2409 of exactly 1.0 for the missing frames, resulting in a smooth "fade in" and
2410 "fade out" at the beginning and at the end of the input, respectively.
2413 Set the compress factor. In range from 0.0 to 30.0. Default is 0.0.
2414 By default, the Dynamic Audio Normalizer does not apply "traditional"
2415 compression. This means that signal peaks will not be pruned and thus the
2416 full dynamic range will be retained within each local neighbourhood. However,
2417 in some cases it may be desirable to combine the Dynamic Audio Normalizer's
2418 normalization algorithm with a more "traditional" compression.
2419 For this purpose, the Dynamic Audio Normalizer provides an optional compression
2420 (thresholding) function. If (and only if) the compression feature is enabled,
2421 all input frames will be processed by a soft knee thresholding function prior
2422 to the actual normalization process. Put simply, the thresholding function is
2423 going to prune all samples whose magnitude exceeds a certain threshold value.
2424 However, the Dynamic Audio Normalizer does not simply apply a fixed threshold
2425 value. Instead, the threshold value will be adjusted for each individual
2427 In general, smaller parameters result in stronger compression, and vice versa.
2428 Values below 3.0 are not recommended, because audible distortion may appear.
2433 Make audio easier to listen to on headphones.
2435 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
2436 so that when listened to on headphones the stereo image is moved from
2437 inside your head (standard for headphones) to outside and in front of
2438 the listener (standard for speakers).
2444 Apply a two-pole peaking equalisation (EQ) filter. With this
2445 filter, the signal-level at and around a selected frequency can
2446 be increased or decreased, whilst (unlike bandpass and bandreject
2447 filters) that at all other frequencies is unchanged.
2449 In order to produce complex equalisation curves, this filter can
2450 be given several times, each with a different central frequency.
2452 The filter accepts the following options:
2456 Set the filter's central frequency in Hz.
2459 Set method to specify band-width of filter.
2472 Specify the band-width of a filter in width_type units.
2475 Set the required gain or attenuation in dB.
2476 Beware of clipping when using a positive gain.
2479 Specify which channels to filter, by default all available are filtered.
2482 @subsection Examples
2485 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
2487 equalizer=f=1000:width_type=h:width=200:g=-10
2491 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
2493 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
2497 @section extrastereo
2499 Linearly increases the difference between left and right channels which
2500 adds some sort of "live" effect to playback.
2502 The filter accepts the following options:
2506 Sets the difference coefficient (default: 2.5). 0.0 means mono sound
2507 (average of both channels), with 1.0 sound will be unchanged, with
2508 -1.0 left and right channels will be swapped.
2511 Enable clipping. By default is enabled.
2514 @section firequalizer
2515 Apply FIR Equalization using arbitrary frequency response.
2517 The filter accepts the following option:
2521 Set gain curve equation (in dB). The expression can contain variables:
2524 the evaluated frequency
2528 channel number, set to 0 when multichannels evaluation is disabled
2530 channel id, see libavutil/channel_layout.h, set to the first channel id when
2531 multichannels evaluation is disabled
2535 channel_layout, see libavutil/channel_layout.h
2540 @item gain_interpolate(f)
2541 interpolate gain on frequency f based on gain_entry
2542 @item cubic_interpolate(f)
2543 same as gain_interpolate, but smoother
2545 This option is also available as command. Default is @code{gain_interpolate(f)}.
2548 Set gain entry for gain_interpolate function. The expression can
2552 store gain entry at frequency f with value g
2554 This option is also available as command.
2557 Set filter delay in seconds. Higher value means more accurate.
2558 Default is @code{0.01}.
2561 Set filter accuracy in Hz. Lower value means more accurate.
2562 Default is @code{5}.
2565 Set window function. Acceptable values are:
2568 rectangular window, useful when gain curve is already smooth
2570 hann window (default)
2576 3-terms continuous 1st derivative nuttall window
2578 minimum 3-terms discontinuous nuttall window
2580 4-terms continuous 1st derivative nuttall window
2582 minimum 4-terms discontinuous nuttall (blackman-nuttall) window
2584 blackman-harris window
2590 If enabled, use fixed number of audio samples. This improves speed when
2591 filtering with large delay. Default is disabled.
2594 Enable multichannels evaluation on gain. Default is disabled.
2597 Enable zero phase mode by subtracting timestamp to compensate delay.
2598 Default is disabled.
2601 Set scale used by gain. Acceptable values are:
2604 linear frequency, linear gain
2606 linear frequency, logarithmic (in dB) gain (default)
2608 logarithmic (in octave scale where 20 Hz is 0) frequency, linear gain
2610 logarithmic frequency, logarithmic gain
2614 Set file for dumping, suitable for gnuplot.
2617 Set scale for dumpfile. Acceptable values are same with scale option.
2621 Enable 2-channel convolution using complex FFT. This improves speed significantly.
2622 Default is disabled.
2625 @subsection Examples
2630 firequalizer=gain='if(lt(f,1000), 0, -INF)'
2633 lowpass at 1000 Hz with gain_entry:
2635 firequalizer=gain_entry='entry(1000,0); entry(1001, -INF)'
2638 custom equalization:
2640 firequalizer=gain_entry='entry(100,0); entry(400, -4); entry(1000, -6); entry(2000, 0)'
2643 higher delay with zero phase to compensate delay:
2645 firequalizer=delay=0.1:fixed=on:zero_phase=on
2648 lowpass on left channel, highpass on right channel:
2650 firequalizer=gain='if(eq(chid,1), gain_interpolate(f), if(eq(chid,2), gain_interpolate(1e6+f), 0))'
2651 :gain_entry='entry(1000, 0); entry(1001,-INF); entry(1e6+1000,0)':multi=on
2656 Apply a flanging effect to the audio.
2658 The filter accepts the following options:
2662 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
2665 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
2668 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
2672 Set percentage of delayed signal mixed with original. Range from 0 to 100.
2673 Default value is 71.
2676 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
2679 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
2680 Default value is @var{sinusoidal}.
2683 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
2684 Default value is 25.
2687 Set delay-line interpolation, @var{linear} or @var{quadratic}.
2688 Default is @var{linear}.
2693 Decodes High Definition Compatible Digital (HDCD) data. A 16-bit PCM stream with
2694 embedded HDCD codes is expanded into a 20-bit PCM stream.
2696 The filter supports the Peak Extend and Low-level Gain Adjustment features
2697 of HDCD, and detects the Transient Filter flag.
2700 ffmpeg -i HDCD16.flac -af hdcd OUT24.flac
2703 When using the filter with wav, note the default encoding for wav is 16-bit,
2704 so the resulting 20-bit stream will be truncated back to 16-bit. Use something
2705 like @command{-acodec pcm_s24le} after the filter to get 24-bit PCM output.
2707 ffmpeg -i HDCD16.wav -af hdcd OUT16.wav
2708 ffmpeg -i HDCD16.wav -af hdcd -acodec pcm_s24le OUT24.wav
2711 The filter accepts the following options:
2714 @item disable_autoconvert
2715 Disable any automatic format conversion or resampling in the filter graph.
2717 @item process_stereo
2718 Process the stereo channels together. If target_gain does not match between
2719 channels, consider it invalid and use the last valid target_gain.
2722 Set the code detect timer period in ms.
2725 Always extend peaks above -3dBFS even if PE isn't signaled.
2728 Replace audio with a solid tone and adjust the amplitude to signal some
2729 specific aspect of the decoding process. The output file can be loaded in
2730 an audio editor alongside the original to aid analysis.
2732 @code{analyze_mode=pe:force_pe=true} can be used to see all samples above the PE level.
2739 Gain adjustment level at each sample
2741 Samples where peak extend occurs
2743 Samples where the code detect timer is active
2745 Samples where the target gain does not match between channels
2751 Apply a high-pass filter with 3dB point frequency.
2752 The filter can be either single-pole, or double-pole (the default).
2753 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2755 The filter accepts the following options:
2759 Set frequency in Hz. Default is 3000.
2762 Set number of poles. Default is 2.
2765 Set method to specify band-width of filter.
2778 Specify the band-width of a filter in width_type units.
2779 Applies only to double-pole filter.
2780 The default is 0.707q and gives a Butterworth response.
2783 Specify which channels to filter, by default all available are filtered.
2788 Join multiple input streams into one multi-channel stream.
2790 It accepts the following parameters:
2794 The number of input streams. It defaults to 2.
2796 @item channel_layout
2797 The desired output channel layout. It defaults to stereo.
2800 Map channels from inputs to output. The argument is a '|'-separated list of
2801 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
2802 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
2803 can be either the name of the input channel (e.g. FL for front left) or its
2804 index in the specified input stream. @var{out_channel} is the name of the output
2808 The filter will attempt to guess the mappings when they are not specified
2809 explicitly. It does so by first trying to find an unused matching input channel
2810 and if that fails it picks the first unused input channel.
2812 Join 3 inputs (with properly set channel layouts):
2814 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
2817 Build a 5.1 output from 6 single-channel streams:
2819 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
2820 'join=inputs=6:channel_layout=5.1:map=0.0-FL|1.0-FR|2.0-FC|3.0-SL|4.0-SR|5.0-LFE'
2826 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
2828 To enable compilation of this filter you need to configure FFmpeg with
2829 @code{--enable-ladspa}.
2833 Specifies the name of LADSPA plugin library to load. If the environment
2834 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
2835 each one of the directories specified by the colon separated list in
2836 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
2837 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
2838 @file{/usr/lib/ladspa/}.
2841 Specifies the plugin within the library. Some libraries contain only
2842 one plugin, but others contain many of them. If this is not set filter
2843 will list all available plugins within the specified library.
2846 Set the '|' separated list of controls which are zero or more floating point
2847 values that determine the behavior of the loaded plugin (for example delay,
2849 Controls need to be defined using the following syntax:
2850 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
2851 @var{valuei} is the value set on the @var{i}-th control.
2852 Alternatively they can be also defined using the following syntax:
2853 @var{value0}|@var{value1}|@var{value2}|..., where
2854 @var{valuei} is the value set on the @var{i}-th control.
2855 If @option{controls} is set to @code{help}, all available controls and
2856 their valid ranges are printed.
2858 @item sample_rate, s
2859 Specify the sample rate, default to 44100. Only used if plugin have
2863 Set the number of samples per channel per each output frame, default
2864 is 1024. Only used if plugin have zero inputs.
2867 Set the minimum duration of the sourced audio. See
2868 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2869 for the accepted syntax.
2870 Note that the resulting duration may be greater than the specified duration,
2871 as the generated audio is always cut at the end of a complete frame.
2872 If not specified, or the expressed duration is negative, the audio is
2873 supposed to be generated forever.
2874 Only used if plugin have zero inputs.
2878 @subsection Examples
2882 List all available plugins within amp (LADSPA example plugin) library:
2888 List all available controls and their valid ranges for @code{vcf_notch}
2889 plugin from @code{VCF} library:
2891 ladspa=f=vcf:p=vcf_notch:c=help
2895 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
2898 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
2902 Add reverberation to the audio using TAP-plugins
2903 (Tom's Audio Processing plugins):
2905 ladspa=file=tap_reverb:tap_reverb
2909 Generate white noise, with 0.2 amplitude:
2911 ladspa=file=cmt:noise_source_white:c=c0=.2
2915 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
2916 @code{C* Audio Plugin Suite} (CAPS) library:
2918 ladspa=file=caps:Click:c=c1=20'
2922 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
2924 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
2928 Increase volume by 20dB using fast lookahead limiter from Steve Harris
2929 @code{SWH Plugins} collection:
2931 ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
2935 Attenuate low frequencies using Multiband EQ from Steve Harris
2936 @code{SWH Plugins} collection:
2938 ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
2942 @subsection Commands
2944 This filter supports the following commands:
2947 Modify the @var{N}-th control value.
2949 If the specified value is not valid, it is ignored and prior one is kept.
2954 EBU R128 loudness normalization. Includes both dynamic and linear normalization modes.
2955 Support for both single pass (livestreams, files) and double pass (files) modes.
2956 This algorithm can target IL, LRA, and maximum true peak.
2958 The filter accepts the following options:
2962 Set integrated loudness target.
2963 Range is -70.0 - -5.0. Default value is -24.0.
2966 Set loudness range target.
2967 Range is 1.0 - 20.0. Default value is 7.0.
2970 Set maximum true peak.
2971 Range is -9.0 - +0.0. Default value is -2.0.
2973 @item measured_I, measured_i
2974 Measured IL of input file.
2975 Range is -99.0 - +0.0.
2977 @item measured_LRA, measured_lra
2978 Measured LRA of input file.
2979 Range is 0.0 - 99.0.
2981 @item measured_TP, measured_tp
2982 Measured true peak of input file.
2983 Range is -99.0 - +99.0.
2985 @item measured_thresh
2986 Measured threshold of input file.
2987 Range is -99.0 - +0.0.
2990 Set offset gain. Gain is applied before the true-peak limiter.
2991 Range is -99.0 - +99.0. Default is +0.0.
2994 Normalize linearly if possible.
2995 measured_I, measured_LRA, measured_TP, and measured_thresh must also
2996 to be specified in order to use this mode.
2997 Options are true or false. Default is true.
3000 Treat mono input files as "dual-mono". If a mono file is intended for playback
3001 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
3002 If set to @code{true}, this option will compensate for this effect.
3003 Multi-channel input files are not affected by this option.
3004 Options are true or false. Default is false.
3007 Set print format for stats. Options are summary, json, or none.
3008 Default value is none.
3013 Apply a low-pass filter with 3dB point frequency.
3014 The filter can be either single-pole or double-pole (the default).
3015 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
3017 The filter accepts the following options:
3021 Set frequency in Hz. Default is 500.
3024 Set number of poles. Default is 2.
3027 Set method to specify band-width of filter.
3040 Specify the band-width of a filter in width_type units.
3041 Applies only to double-pole filter.
3042 The default is 0.707q and gives a Butterworth response.
3045 Specify which channels to filter, by default all available are filtered.
3048 @subsection Examples
3051 Lowpass only LFE channel, it LFE is not present it does nothing:
3060 Mix channels with specific gain levels. The filter accepts the output
3061 channel layout followed by a set of channels definitions.
3063 This filter is also designed to efficiently remap the channels of an audio
3066 The filter accepts parameters of the form:
3067 "@var{l}|@var{outdef}|@var{outdef}|..."
3071 output channel layout or number of channels
3074 output channel specification, of the form:
3075 "@var{out_name}=[@var{gain}*]@var{in_name}[(+-)[@var{gain}*]@var{in_name}...]"
3078 output channel to define, either a channel name (FL, FR, etc.) or a channel
3079 number (c0, c1, etc.)
3082 multiplicative coefficient for the channel, 1 leaving the volume unchanged
3085 input channel to use, see out_name for details; it is not possible to mix
3086 named and numbered input channels
3089 If the `=' in a channel specification is replaced by `<', then the gains for
3090 that specification will be renormalized so that the total is 1, thus
3091 avoiding clipping noise.
3093 @subsection Mixing examples
3095 For example, if you want to down-mix from stereo to mono, but with a bigger
3096 factor for the left channel:
3098 pan=1c|c0=0.9*c0+0.1*c1
3101 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
3102 7-channels surround:
3104 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
3107 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
3108 that should be preferred (see "-ac" option) unless you have very specific
3111 @subsection Remapping examples
3113 The channel remapping will be effective if, and only if:
3116 @item gain coefficients are zeroes or ones,
3117 @item only one input per channel output,
3120 If all these conditions are satisfied, the filter will notify the user ("Pure
3121 channel mapping detected"), and use an optimized and lossless method to do the
3124 For example, if you have a 5.1 source and want a stereo audio stream by
3125 dropping the extra channels:
3127 pan="stereo| c0=FL | c1=FR"
3130 Given the same source, you can also switch front left and front right channels
3131 and keep the input channel layout:
3133 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
3136 If the input is a stereo audio stream, you can mute the front left channel (and
3137 still keep the stereo channel layout) with:
3142 Still with a stereo audio stream input, you can copy the right channel in both
3143 front left and right:
3145 pan="stereo| c0=FR | c1=FR"
3150 ReplayGain scanner filter. This filter takes an audio stream as an input and
3151 outputs it unchanged.
3152 At end of filtering it displays @code{track_gain} and @code{track_peak}.
3156 Convert the audio sample format, sample rate and channel layout. It is
3157 not meant to be used directly.
3160 Apply time-stretching and pitch-shifting with librubberband.
3162 The filter accepts the following options:
3166 Set tempo scale factor.
3169 Set pitch scale factor.
3172 Set transients detector.
3173 Possible values are:
3182 Possible values are:
3191 Possible values are:
3198 Set processing window size.
3199 Possible values are:
3208 Possible values are:
3215 Enable formant preservation when shift pitching.
3216 Possible values are:
3224 Possible values are:
3233 Possible values are:
3240 @section sidechaincompress
3242 This filter acts like normal compressor but has the ability to compress
3243 detected signal using second input signal.
3244 It needs two input streams and returns one output stream.
3245 First input stream will be processed depending on second stream signal.
3246 The filtered signal then can be filtered with other filters in later stages of
3247 processing. See @ref{pan} and @ref{amerge} filter.
3249 The filter accepts the following options:
3253 Set input gain. Default is 1. Range is between 0.015625 and 64.
3256 If a signal of second stream raises above this level it will affect the gain
3257 reduction of first stream.
3258 By default is 0.125. Range is between 0.00097563 and 1.
3261 Set a ratio about which the signal is reduced. 1:2 means that if the level
3262 raised 4dB above the threshold, it will be only 2dB above after the reduction.
3263 Default is 2. Range is between 1 and 20.
3266 Amount of milliseconds the signal has to rise above the threshold before gain
3267 reduction starts. Default is 20. Range is between 0.01 and 2000.
3270 Amount of milliseconds the signal has to fall below the threshold before
3271 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
3274 Set the amount by how much signal will be amplified after processing.
3275 Default is 2. Range is from 1 and 64.
3278 Curve the sharp knee around the threshold to enter gain reduction more softly.
3279 Default is 2.82843. Range is between 1 and 8.
3282 Choose if the @code{average} level between all channels of side-chain stream
3283 or the louder(@code{maximum}) channel of side-chain stream affects the
3284 reduction. Default is @code{average}.
3287 Should the exact signal be taken in case of @code{peak} or an RMS one in case
3288 of @code{rms}. Default is @code{rms} which is mainly smoother.
3291 Set sidechain gain. Default is 1. Range is between 0.015625 and 64.
3294 How much to use compressed signal in output. Default is 1.
3295 Range is between 0 and 1.
3298 @subsection Examples
3302 Full ffmpeg example taking 2 audio inputs, 1st input to be compressed
3303 depending on the signal of 2nd input and later compressed signal to be
3304 merged with 2nd input:
3306 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
3310 @section sidechaingate
3312 A sidechain gate acts like a normal (wideband) gate but has the ability to
3313 filter the detected signal before sending it to the gain reduction stage.
3314 Normally a gate uses the full range signal to detect a level above the
3316 For example: If you cut all lower frequencies from your sidechain signal
3317 the gate will decrease the volume of your track only if not enough highs
3318 appear. With this technique you are able to reduce the resonation of a
3319 natural drum or remove "rumbling" of muted strokes from a heavily distorted
3321 It needs two input streams and returns one output stream.
3322 First input stream will be processed depending on second stream signal.
3324 The filter accepts the following options:
3328 Set input level before filtering.
3329 Default is 1. Allowed range is from 0.015625 to 64.
3332 Set the level of gain reduction when the signal is below the threshold.
3333 Default is 0.06125. Allowed range is from 0 to 1.
3336 If a signal rises above this level the gain reduction is released.
3337 Default is 0.125. Allowed range is from 0 to 1.
3340 Set a ratio about which the signal is reduced.
3341 Default is 2. Allowed range is from 1 to 9000.
3344 Amount of milliseconds the signal has to rise above the threshold before gain
3346 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
3349 Amount of milliseconds the signal has to fall below the threshold before the
3350 reduction is increased again. Default is 250 milliseconds.
3351 Allowed range is from 0.01 to 9000.
3354 Set amount of amplification of signal after processing.
3355 Default is 1. Allowed range is from 1 to 64.
3358 Curve the sharp knee around the threshold to enter gain reduction more softly.
3359 Default is 2.828427125. Allowed range is from 1 to 8.
3362 Choose if exact signal should be taken for detection or an RMS like one.
3363 Default is rms. Can be peak or rms.
3366 Choose if the average level between all channels or the louder channel affects
3368 Default is average. Can be average or maximum.
3371 Set sidechain gain. Default is 1. Range is from 0.015625 to 64.
3374 @section silencedetect
3376 Detect silence in an audio stream.
3378 This filter logs a message when it detects that the input audio volume is less
3379 or equal to a noise tolerance value for a duration greater or equal to the
3380 minimum detected noise duration.
3382 The printed times and duration are expressed in seconds.
3384 The filter accepts the following options:
3388 Set silence duration until notification (default is 2 seconds).
3391 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
3392 specified value) or amplitude ratio. Default is -60dB, or 0.001.
3395 @subsection Examples
3399 Detect 5 seconds of silence with -50dB noise tolerance:
3401 silencedetect=n=-50dB:d=5
3405 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
3406 tolerance in @file{silence.mp3}:
3408 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
3412 @section silenceremove
3414 Remove silence from the beginning, middle or end of the audio.
3416 The filter accepts the following options:
3420 This value is used to indicate if audio should be trimmed at beginning of
3421 the audio. A value of zero indicates no silence should be trimmed from the
3422 beginning. When specifying a non-zero value, it trims audio up until it
3423 finds non-silence. Normally, when trimming silence from beginning of audio
3424 the @var{start_periods} will be @code{1} but it can be increased to higher
3425 values to trim all audio up to specific count of non-silence periods.
3426 Default value is @code{0}.
3428 @item start_duration
3429 Specify the amount of time that non-silence must be detected before it stops
3430 trimming audio. By increasing the duration, bursts of noises can be treated
3431 as silence and trimmed off. Default value is @code{0}.
3433 @item start_threshold
3434 This indicates what sample value should be treated as silence. For digital
3435 audio, a value of @code{0} may be fine but for audio recorded from analog,
3436 you may wish to increase the value to account for background noise.
3437 Can be specified in dB (in case "dB" is appended to the specified value)
3438 or amplitude ratio. Default value is @code{0}.
3441 Set the count for trimming silence from the end of audio.
3442 To remove silence from the middle of a file, specify a @var{stop_periods}
3443 that is negative. This value is then treated as a positive value and is
3444 used to indicate the effect should restart processing as specified by
3445 @var{start_periods}, making it suitable for removing periods of silence
3446 in the middle of the audio.
3447 Default value is @code{0}.
3450 Specify a duration of silence that must exist before audio is not copied any
3451 more. By specifying a higher duration, silence that is wanted can be left in
3453 Default value is @code{0}.
3455 @item stop_threshold
3456 This is the same as @option{start_threshold} but for trimming silence from
3458 Can be specified in dB (in case "dB" is appended to the specified value)
3459 or amplitude ratio. Default value is @code{0}.
3462 This indicates that @var{stop_duration} length of audio should be left intact
3463 at the beginning of each period of silence.
3464 For example, if you want to remove long pauses between words but do not want
3465 to remove the pauses completely. Default value is @code{0}.
3468 Set how is silence detected. Can be @code{rms} or @code{peak}. Second is faster
3469 and works better with digital silence which is exactly 0.
3470 Default value is @code{rms}.
3473 Set ratio used to calculate size of window for detecting silence.
3474 Default value is @code{0.02}. Allowed range is from @code{0} to @code{10}.
3477 @subsection Examples
3481 The following example shows how this filter can be used to start a recording
3482 that does not contain the delay at the start which usually occurs between
3483 pressing the record button and the start of the performance:
3485 silenceremove=1:5:0.02
3489 Trim all silence encountered from beginning to end where there is more than 1
3490 second of silence in audio:
3492 silenceremove=0:0:0:-1:1:-90dB
3498 SOFAlizer uses head-related transfer functions (HRTFs) to create virtual
3499 loudspeakers around the user for binaural listening via headphones (audio
3500 formats up to 9 channels supported).
3501 The HRTFs are stored in SOFA files (see @url{http://www.sofacoustics.org/} for a database).
3502 SOFAlizer is developed at the Acoustics Research Institute (ARI) of the
3503 Austrian Academy of Sciences.
3505 To enable compilation of this filter you need to configure FFmpeg with
3506 @code{--enable-netcdf}.
3508 The filter accepts the following options:
3512 Set the SOFA file used for rendering.
3515 Set gain applied to audio. Value is in dB. Default is 0.
3518 Set rotation of virtual loudspeakers in deg. Default is 0.
3521 Set elevation of virtual speakers in deg. Default is 0.
3524 Set distance in meters between loudspeakers and the listener with near-field
3525 HRTFs. Default is 1.
3528 Set processing type. Can be @var{time} or @var{freq}. @var{time} is
3529 processing audio in time domain which is slow.
3530 @var{freq} is processing audio in frequency domain which is fast.
3531 Default is @var{freq}.
3534 Set custom positions of virtual loudspeakers. Syntax for this option is:
3535 <CH> <AZIM> <ELEV>[|<CH> <AZIM> <ELEV>|...].
3536 Each virtual loudspeaker is described with short channel name following with
3537 azimuth and elevation in degreees.
3538 Each virtual loudspeaker description is separated by '|'.
3539 For example to override front left and front right channel positions use:
3540 'speakers=FL 45 15|FR 345 15'.
3541 Descriptions with unrecognised channel names are ignored.
3544 @subsection Examples
3548 Using ClubFritz6 sofa file:
3550 sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=1
3554 Using ClubFritz12 sofa file and bigger radius with small rotation:
3556 sofalizer=sofa=/path/to/ClubFritz12.sofa:type=freq:radius=2:rotation=5
3560 Similar as above but with custom speaker positions for front left, front right, back left and back right
3561 and also with custom gain:
3563 "sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=2:speakers=FL 45|FR 315|BL 135|BR 225:gain=28"
3567 @section stereotools
3569 This filter has some handy utilities to manage stereo signals, for converting
3570 M/S stereo recordings to L/R signal while having control over the parameters
3571 or spreading the stereo image of master track.
3573 The filter accepts the following options:
3577 Set input level before filtering for both channels. Defaults is 1.
3578 Allowed range is from 0.015625 to 64.
3581 Set output level after filtering for both channels. Defaults is 1.
3582 Allowed range is from 0.015625 to 64.
3585 Set input balance between both channels. Default is 0.
3586 Allowed range is from -1 to 1.
3589 Set output balance between both channels. Default is 0.
3590 Allowed range is from -1 to 1.
3593 Enable softclipping. Results in analog distortion instead of harsh digital 0dB
3594 clipping. Disabled by default.
3597 Mute the left channel. Disabled by default.
3600 Mute the right channel. Disabled by default.
3603 Change the phase of the left channel. Disabled by default.
3606 Change the phase of the right channel. Disabled by default.
3609 Set stereo mode. Available values are:
3613 Left/Right to Left/Right, this is default.
3616 Left/Right to Mid/Side.
3619 Mid/Side to Left/Right.
3622 Left/Right to Left/Left.
3625 Left/Right to Right/Right.
3628 Left/Right to Left + Right.
3631 Left/Right to Right/Left.
3635 Set level of side signal. Default is 1.
3636 Allowed range is from 0.015625 to 64.
3639 Set balance of side signal. Default is 0.
3640 Allowed range is from -1 to 1.
3643 Set level of the middle signal. Default is 1.
3644 Allowed range is from 0.015625 to 64.
3647 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
3650 Set stereo base between mono and inversed channels. Default is 0.
3651 Allowed range is from -1 to 1.
3654 Set delay in milliseconds how much to delay left from right channel and
3655 vice versa. Default is 0. Allowed range is from -20 to 20.
3658 Set S/C level. Default is 1. Allowed range is from 1 to 100.
3661 Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
3664 @subsection Examples
3668 Apply karaoke like effect:
3670 stereotools=mlev=0.015625
3674 Convert M/S signal to L/R:
3676 "stereotools=mode=ms>lr"
3680 @section stereowiden
3682 This filter enhance the stereo effect by suppressing signal common to both
3683 channels and by delaying the signal of left into right and vice versa,
3684 thereby widening the stereo effect.
3686 The filter accepts the following options:
3690 Time in milliseconds of the delay of left signal into right and vice versa.
3691 Default is 20 milliseconds.
3694 Amount of gain in delayed signal into right and vice versa. Gives a delay
3695 effect of left signal in right output and vice versa which gives widening
3696 effect. Default is 0.3.
3699 Cross feed of left into right with inverted phase. This helps in suppressing
3700 the mono. If the value is 1 it will cancel all the signal common to both
3701 channels. Default is 0.3.
3704 Set level of input signal of original channel. Default is 0.8.
3709 Boost or cut treble (upper) frequencies of the audio using a two-pole
3710 shelving filter with a response similar to that of a standard
3711 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
3713 The filter accepts the following options:
3717 Give the gain at whichever is the lower of ~22 kHz and the
3718 Nyquist frequency. Its useful range is about -20 (for a large cut)
3719 to +20 (for a large boost). Beware of clipping when using a positive gain.
3722 Set the filter's central frequency and so can be used
3723 to extend or reduce the frequency range to be boosted or cut.
3724 The default value is @code{3000} Hz.
3727 Set method to specify band-width of filter.
3740 Determine how steep is the filter's shelf transition.
3743 Specify which channels to filter, by default all available are filtered.
3748 Sinusoidal amplitude modulation.
3750 The filter accepts the following options:
3754 Modulation frequency in Hertz. Modulation frequencies in the subharmonic range
3755 (20 Hz or lower) will result in a tremolo effect.
3756 This filter may also be used as a ring modulator by specifying
3757 a modulation frequency higher than 20 Hz.
3758 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3761 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3762 Default value is 0.5.
3767 Sinusoidal phase modulation.
3769 The filter accepts the following options:
3773 Modulation frequency in Hertz.
3774 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3777 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3778 Default value is 0.5.
3783 Adjust the input audio volume.
3785 It accepts the following parameters:
3789 Set audio volume expression.
3791 Output values are clipped to the maximum value.
3793 The output audio volume is given by the relation:
3795 @var{output_volume} = @var{volume} * @var{input_volume}
3798 The default value for @var{volume} is "1.0".
3801 This parameter represents the mathematical precision.
3803 It determines which input sample formats will be allowed, which affects the
3804 precision of the volume scaling.
3808 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
3810 32-bit floating-point; this limits input sample format to FLT. (default)
3812 64-bit floating-point; this limits input sample format to DBL.
3816 Choose the behaviour on encountering ReplayGain side data in input frames.
3820 Remove ReplayGain side data, ignoring its contents (the default).
3823 Ignore ReplayGain side data, but leave it in the frame.
3826 Prefer the track gain, if present.
3829 Prefer the album gain, if present.
3832 @item replaygain_preamp
3833 Pre-amplification gain in dB to apply to the selected replaygain gain.
3835 Default value for @var{replaygain_preamp} is 0.0.
3838 Set when the volume expression is evaluated.
3840 It accepts the following values:
3843 only evaluate expression once during the filter initialization, or
3844 when the @samp{volume} command is sent
3847 evaluate expression for each incoming frame
3850 Default value is @samp{once}.
3853 The volume expression can contain the following parameters.
3857 frame number (starting at zero)
3860 @item nb_consumed_samples
3861 number of samples consumed by the filter
3863 number of samples in the current frame
3865 original frame position in the file
3871 PTS at start of stream
3873 time at start of stream
3879 last set volume value
3882 Note that when @option{eval} is set to @samp{once} only the
3883 @var{sample_rate} and @var{tb} variables are available, all other
3884 variables will evaluate to NAN.
3886 @subsection Commands
3888 This filter supports the following commands:
3891 Modify the volume expression.
3892 The command accepts the same syntax of the corresponding option.
3894 If the specified expression is not valid, it is kept at its current
3896 @item replaygain_noclip
3897 Prevent clipping by limiting the gain applied.
3899 Default value for @var{replaygain_noclip} is 1.
3903 @subsection Examples
3907 Halve the input audio volume:
3911 volume=volume=-6.0206dB
3914 In all the above example the named key for @option{volume} can be
3915 omitted, for example like in:
3921 Increase input audio power by 6 decibels using fixed-point precision:
3923 volume=volume=6dB:precision=fixed
3927 Fade volume after time 10 with an annihilation period of 5 seconds:
3929 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
3933 @section volumedetect
3935 Detect the volume of the input video.
3937 The filter has no parameters. The input is not modified. Statistics about
3938 the volume will be printed in the log when the input stream end is reached.
3940 In particular it will show the mean volume (root mean square), maximum
3941 volume (on a per-sample basis), and the beginning of a histogram of the
3942 registered volume values (from the maximum value to a cumulated 1/1000 of
3945 All volumes are in decibels relative to the maximum PCM value.
3947 @subsection Examples
3949 Here is an excerpt of the output:
3951 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
3952 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
3953 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
3954 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
3955 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
3956 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
3957 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
3958 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
3959 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
3965 The mean square energy is approximately -27 dB, or 10^-2.7.
3967 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
3969 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
3972 In other words, raising the volume by +4 dB does not cause any clipping,
3973 raising it by +5 dB causes clipping for 6 samples, etc.
3975 @c man end AUDIO FILTERS
3977 @chapter Audio Sources
3978 @c man begin AUDIO SOURCES
3980 Below is a description of the currently available audio sources.
3984 Buffer audio frames, and make them available to the filter chain.
3986 This source is mainly intended for a programmatic use, in particular
3987 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
3989 It accepts the following parameters:
3993 The timebase which will be used for timestamps of submitted frames. It must be
3994 either a floating-point number or in @var{numerator}/@var{denominator} form.
3997 The sample rate of the incoming audio buffers.
4000 The sample format of the incoming audio buffers.
4001 Either a sample format name or its corresponding integer representation from
4002 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
4004 @item channel_layout
4005 The channel layout of the incoming audio buffers.
4006 Either a channel layout name from channel_layout_map in
4007 @file{libavutil/channel_layout.c} or its corresponding integer representation
4008 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
4011 The number of channels of the incoming audio buffers.
4012 If both @var{channels} and @var{channel_layout} are specified, then they
4017 @subsection Examples
4020 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
4023 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
4024 Since the sample format with name "s16p" corresponds to the number
4025 6 and the "stereo" channel layout corresponds to the value 0x3, this is
4028 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
4033 Generate an audio signal specified by an expression.
4035 This source accepts in input one or more expressions (one for each
4036 channel), which are evaluated and used to generate a corresponding
4039 This source accepts the following options:
4043 Set the '|'-separated expressions list for each separate channel. In case the
4044 @option{channel_layout} option is not specified, the selected channel layout
4045 depends on the number of provided expressions. Otherwise the last
4046 specified expression is applied to the remaining output channels.
4048 @item channel_layout, c
4049 Set the channel layout. The number of channels in the specified layout
4050 must be equal to the number of specified expressions.
4053 Set the minimum duration of the sourced audio. See
4054 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
4055 for the accepted syntax.
4056 Note that the resulting duration may be greater than the specified
4057 duration, as the generated audio is always cut at the end of a
4060 If not specified, or the expressed duration is negative, the audio is
4061 supposed to be generated forever.
4064 Set the number of samples per channel per each output frame,
4067 @item sample_rate, s
4068 Specify the sample rate, default to 44100.
4071 Each expression in @var{exprs} can contain the following constants:
4075 number of the evaluated sample, starting from 0
4078 time of the evaluated sample expressed in seconds, starting from 0
4085 @subsection Examples
4095 Generate a sin signal with frequency of 440 Hz, set sample rate to
4098 aevalsrc="sin(440*2*PI*t):s=8000"
4102 Generate a two channels signal, specify the channel layout (Front
4103 Center + Back Center) explicitly:
4105 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
4109 Generate white noise:
4111 aevalsrc="-2+random(0)"
4115 Generate an amplitude modulated signal:
4117 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
4121 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
4123 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
4130 The null audio source, return unprocessed audio frames. It is mainly useful
4131 as a template and to be employed in analysis / debugging tools, or as
4132 the source for filters which ignore the input data (for example the sox
4135 This source accepts the following options:
4139 @item channel_layout, cl
4141 Specifies the channel layout, and can be either an integer or a string
4142 representing a channel layout. The default value of @var{channel_layout}
4145 Check the channel_layout_map definition in
4146 @file{libavutil/channel_layout.c} for the mapping between strings and
4147 channel layout values.
4149 @item sample_rate, r
4150 Specifies the sample rate, and defaults to 44100.
4153 Set the number of samples per requested frames.
4157 @subsection Examples
4161 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
4163 anullsrc=r=48000:cl=4
4167 Do the same operation with a more obvious syntax:
4169 anullsrc=r=48000:cl=mono
4173 All the parameters need to be explicitly defined.
4177 Synthesize a voice utterance using the libflite library.
4179 To enable compilation of this filter you need to configure FFmpeg with
4180 @code{--enable-libflite}.
4182 Note that the flite library is not thread-safe.
4184 The filter accepts the following options:
4189 If set to 1, list the names of the available voices and exit
4190 immediately. Default value is 0.
4193 Set the maximum number of samples per frame. Default value is 512.
4196 Set the filename containing the text to speak.
4199 Set the text to speak.
4202 Set the voice to use for the speech synthesis. Default value is
4203 @code{kal}. See also the @var{list_voices} option.
4206 @subsection Examples
4210 Read from file @file{speech.txt}, and synthesize the text using the
4211 standard flite voice:
4213 flite=textfile=speech.txt
4217 Read the specified text selecting the @code{slt} voice:
4219 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4223 Input text to ffmpeg:
4225 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4229 Make @file{ffplay} speak the specified text, using @code{flite} and
4230 the @code{lavfi} device:
4232 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
4236 For more information about libflite, check:
4237 @url{http://www.speech.cs.cmu.edu/flite/}
4241 Generate a noise audio signal.
4243 The filter accepts the following options:
4246 @item sample_rate, r
4247 Specify the sample rate. Default value is 48000 Hz.
4250 Specify the amplitude (0.0 - 1.0) of the generated audio stream. Default value
4254 Specify the duration of the generated audio stream. Not specifying this option
4255 results in noise with an infinite length.
4257 @item color, colour, c
4258 Specify the color of noise. Available noise colors are white, pink, and brown.
4259 Default color is white.
4262 Specify a value used to seed the PRNG.
4265 Set the number of samples per each output frame, default is 1024.
4268 @subsection Examples
4273 Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate and an amplitude of 0.5:
4275 anoisesrc=d=60:c=pink:r=44100:a=0.5
4281 Generate an audio signal made of a sine wave with amplitude 1/8.
4283 The audio signal is bit-exact.
4285 The filter accepts the following options:
4290 Set the carrier frequency. Default is 440 Hz.
4292 @item beep_factor, b
4293 Enable a periodic beep every second with frequency @var{beep_factor} times
4294 the carrier frequency. Default is 0, meaning the beep is disabled.
4296 @item sample_rate, r
4297 Specify the sample rate, default is 44100.
4300 Specify the duration of the generated audio stream.
4302 @item samples_per_frame
4303 Set the number of samples per output frame.
4305 The expression can contain the following constants:
4309 The (sequential) number of the output audio frame, starting from 0.
4312 The PTS (Presentation TimeStamp) of the output audio frame,
4313 expressed in @var{TB} units.
4316 The PTS of the output audio frame, expressed in seconds.
4319 The timebase of the output audio frames.
4322 Default is @code{1024}.
4325 @subsection Examples
4330 Generate a simple 440 Hz sine wave:
4336 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
4340 sine=frequency=220:beep_factor=4:duration=5
4344 Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
4347 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
4351 @c man end AUDIO SOURCES
4353 @chapter Audio Sinks
4354 @c man begin AUDIO SINKS
4356 Below is a description of the currently available audio sinks.
4358 @section abuffersink
4360 Buffer audio frames, and make them available to the end of filter chain.
4362 This sink is mainly intended for programmatic use, in particular
4363 through the interface defined in @file{libavfilter/buffersink.h}
4364 or the options system.
4366 It accepts a pointer to an AVABufferSinkContext structure, which
4367 defines the incoming buffers' formats, to be passed as the opaque
4368 parameter to @code{avfilter_init_filter} for initialization.
4371 Null audio sink; do absolutely nothing with the input audio. It is
4372 mainly useful as a template and for use in analysis / debugging
4375 @c man end AUDIO SINKS
4377 @chapter Video Filters
4378 @c man begin VIDEO FILTERS
4380 When you configure your FFmpeg build, you can disable any of the
4381 existing filters using @code{--disable-filters}.
4382 The configure output will show the video filters included in your
4385 Below is a description of the currently available video filters.
4387 @section alphaextract
4389 Extract the alpha component from the input as a grayscale video. This
4390 is especially useful with the @var{alphamerge} filter.
4394 Add or replace the alpha component of the primary input with the
4395 grayscale value of a second input. This is intended for use with
4396 @var{alphaextract} to allow the transmission or storage of frame
4397 sequences that have alpha in a format that doesn't support an alpha
4400 For example, to reconstruct full frames from a normal YUV-encoded video
4401 and a separate video created with @var{alphaextract}, you might use:
4403 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
4406 Since this filter is designed for reconstruction, it operates on frame
4407 sequences without considering timestamps, and terminates when either
4408 input reaches end of stream. This will cause problems if your encoding
4409 pipeline drops frames. If you're trying to apply an image as an
4410 overlay to a video stream, consider the @var{overlay} filter instead.
4414 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
4415 and libavformat to work. On the other hand, it is limited to ASS (Advanced
4416 Substation Alpha) subtitles files.
4418 This filter accepts the following option in addition to the common options from
4419 the @ref{subtitles} filter:
4423 Set the shaping engine
4425 Available values are:
4428 The default libass shaping engine, which is the best available.
4430 Fast, font-agnostic shaper that can do only substitutions
4432 Slower shaper using OpenType for substitutions and positioning
4435 The default is @code{auto}.
4439 Apply an Adaptive Temporal Averaging Denoiser to the video input.
4441 The filter accepts the following options:
4445 Set threshold A for 1st plane. Default is 0.02.
4446 Valid range is 0 to 0.3.
4449 Set threshold B for 1st plane. Default is 0.04.
4450 Valid range is 0 to 5.
4453 Set threshold A for 2nd plane. Default is 0.02.
4454 Valid range is 0 to 0.3.
4457 Set threshold B for 2nd plane. Default is 0.04.
4458 Valid range is 0 to 5.
4461 Set threshold A for 3rd plane. Default is 0.02.
4462 Valid range is 0 to 0.3.
4465 Set threshold B for 3rd plane. Default is 0.04.
4466 Valid range is 0 to 5.
4468 Threshold A is designed to react on abrupt changes in the input signal and
4469 threshold B is designed to react on continuous changes in the input signal.
4472 Set number of frames filter will use for averaging. Default is 33. Must be odd
4473 number in range [5, 129].
4476 Set what planes of frame filter will use for averaging. Default is all.
4481 Apply average blur filter.
4483 The filter accepts the following options:
4487 Set horizontal kernel size.
4490 Set which planes to filter. By default all planes are filtered.
4493 Set vertical kernel size, if zero it will be same as @code{sizeX}.
4494 Default is @code{0}.
4499 Compute the bounding box for the non-black pixels in the input frame
4502 This filter computes the bounding box containing all the pixels with a
4503 luminance value greater than the minimum allowed value.
4504 The parameters describing the bounding box are printed on the filter
4507 The filter accepts the following option:
4511 Set the minimal luminance value. Default is @code{16}.
4514 @section bitplanenoise
4516 Show and measure bit plane noise.
4518 The filter accepts the following options:
4522 Set which plane to analyze. Default is @code{1}.
4525 Filter out noisy pixels from @code{bitplane} set above.
4526 Default is disabled.
4529 @section blackdetect
4531 Detect video intervals that are (almost) completely black. Can be
4532 useful to detect chapter transitions, commercials, or invalid
4533 recordings. Output lines contains the time for the start, end and
4534 duration of the detected black interval expressed in seconds.
4536 In order to display the output lines, you need to set the loglevel at
4537 least to the AV_LOG_INFO value.
4539 The filter accepts the following options:
4542 @item black_min_duration, d
4543 Set the minimum detected black duration expressed in seconds. It must
4544 be a non-negative floating point number.
4546 Default value is 2.0.
4548 @item picture_black_ratio_th, pic_th
4549 Set the threshold for considering a picture "black".
4550 Express the minimum value for the ratio:
4552 @var{nb_black_pixels} / @var{nb_pixels}
4555 for which a picture is considered black.
4556 Default value is 0.98.
4558 @item pixel_black_th, pix_th
4559 Set the threshold for considering a pixel "black".
4561 The threshold expresses the maximum pixel luminance value for which a
4562 pixel is considered "black". The provided value is scaled according to
4563 the following equation:
4565 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
4568 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
4569 the input video format, the range is [0-255] for YUV full-range
4570 formats and [16-235] for YUV non full-range formats.
4572 Default value is 0.10.
4575 The following example sets the maximum pixel threshold to the minimum
4576 value, and detects only black intervals of 2 or more seconds:
4578 blackdetect=d=2:pix_th=0.00
4583 Detect frames that are (almost) completely black. Can be useful to
4584 detect chapter transitions or commercials. Output lines consist of
4585 the frame number of the detected frame, the percentage of blackness,
4586 the position in the file if known or -1 and the timestamp in seconds.
4588 In order to display the output lines, you need to set the loglevel at
4589 least to the AV_LOG_INFO value.
4591 This filter exports frame metadata @code{lavfi.blackframe.pblack}.
4592 The value represents the percentage of pixels in the picture that
4593 are below the threshold value.
4595 It accepts the following parameters:
4600 The percentage of the pixels that have to be below the threshold; it defaults to
4603 @item threshold, thresh
4604 The threshold below which a pixel value is considered black; it defaults to
4609 @section blend, tblend
4611 Blend two video frames into each other.
4613 The @code{blend} filter takes two input streams and outputs one
4614 stream, the first input is the "top" layer and second input is
4615 "bottom" layer. By default, the output terminates when the longest input terminates.
4617 The @code{tblend} (time blend) filter takes two consecutive frames
4618 from one single stream, and outputs the result obtained by blending
4619 the new frame on top of the old frame.
4621 A description of the accepted options follows.
4629 Set blend mode for specific pixel component or all pixel components in case
4630 of @var{all_mode}. Default value is @code{normal}.
4632 Available values for component modes are:
4673 Set blend opacity for specific pixel component or all pixel components in case
4674 of @var{all_opacity}. Only used in combination with pixel component blend modes.
4681 Set blend expression for specific pixel component or all pixel components in case
4682 of @var{all_expr}. Note that related mode options will be ignored if those are set.
4684 The expressions can use the following variables:
4688 The sequential number of the filtered frame, starting from @code{0}.
4692 the coordinates of the current sample
4696 the width and height of currently filtered plane
4700 Width and height scale depending on the currently filtered plane. It is the
4701 ratio between the corresponding luma plane number of pixels and the current
4702 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
4703 @code{0.5,0.5} for chroma planes.
4706 Time of the current frame, expressed in seconds.
4709 Value of pixel component at current location for first video frame (top layer).
4712 Value of pixel component at current location for second video frame (bottom layer).
4716 Force termination when the shortest input terminates. Default is
4717 @code{0}. This option is only defined for the @code{blend} filter.
4720 Continue applying the last bottom frame after the end of the stream. A value of
4721 @code{0} disable the filter after the last frame of the bottom layer is reached.
4722 Default is @code{1}. This option is only defined for the @code{blend} filter.
4725 @subsection Examples
4729 Apply transition from bottom layer to top layer in first 10 seconds:
4731 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
4735 Apply 1x1 checkerboard effect:
4737 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
4741 Apply uncover left effect:
4743 blend=all_expr='if(gte(N*SW+X,W),A,B)'
4747 Apply uncover down effect:
4749 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
4753 Apply uncover up-left effect:
4755 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
4759 Split diagonally video and shows top and bottom layer on each side:
4761 blend=all_expr=if(gt(X,Y*(W/H)),A,B)
4765 Display differences between the current and the previous frame:
4767 tblend=all_mode=difference128
4773 Apply a boxblur algorithm to the input video.
4775 It accepts the following parameters:
4779 @item luma_radius, lr
4780 @item luma_power, lp
4781 @item chroma_radius, cr
4782 @item chroma_power, cp
4783 @item alpha_radius, ar
4784 @item alpha_power, ap
4788 A description of the accepted options follows.
4791 @item luma_radius, lr
4792 @item chroma_radius, cr
4793 @item alpha_radius, ar
4794 Set an expression for the box radius in pixels used for blurring the
4795 corresponding input plane.
4797 The radius value must be a non-negative number, and must not be
4798 greater than the value of the expression @code{min(w,h)/2} for the
4799 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
4802 Default value for @option{luma_radius} is "2". If not specified,
4803 @option{chroma_radius} and @option{alpha_radius} default to the
4804 corresponding value set for @option{luma_radius}.
4806 The expressions can contain the following constants:
4810 The input width and height in pixels.
4814 The input chroma image width and height in pixels.
4818 The horizontal and vertical chroma subsample values. For example, for the
4819 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
4822 @item luma_power, lp
4823 @item chroma_power, cp
4824 @item alpha_power, ap
4825 Specify how many times the boxblur filter is applied to the
4826 corresponding plane.
4828 Default value for @option{luma_power} is 2. If not specified,
4829 @option{chroma_power} and @option{alpha_power} default to the
4830 corresponding value set for @option{luma_power}.
4832 A value of 0 will disable the effect.
4835 @subsection Examples
4839 Apply a boxblur filter with the luma, chroma, and alpha radii
4842 boxblur=luma_radius=2:luma_power=1
4847 Set the luma radius to 2, and alpha and chroma radius to 0:
4849 boxblur=2:1:cr=0:ar=0
4853 Set the luma and chroma radii to a fraction of the video dimension:
4855 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
4861 Deinterlace the input video ("bwdif" stands for "Bob Weaver
4862 Deinterlacing Filter").
4864 Motion adaptive deinterlacing based on yadif with the use of w3fdif and cubic
4865 interpolation algorithms.
4866 It accepts the following parameters:
4870 The interlacing mode to adopt. It accepts one of the following values:
4874 Output one frame for each frame.
4876 Output one frame for each field.
4879 The default value is @code{send_field}.
4882 The picture field parity assumed for the input interlaced video. It accepts one
4883 of the following values:
4887 Assume the top field is first.
4889 Assume the bottom field is first.
4891 Enable automatic detection of field parity.
4894 The default value is @code{auto}.
4895 If the interlacing is unknown or the decoder does not export this information,
4896 top field first will be assumed.
4899 Specify which frames to deinterlace. Accept one of the following
4904 Deinterlace all frames.
4906 Only deinterlace frames marked as interlaced.
4909 The default value is @code{all}.
4913 YUV colorspace color/chroma keying.
4915 The filter accepts the following options:
4919 The color which will be replaced with transparency.
4922 Similarity percentage with the key color.
4924 0.01 matches only the exact key color, while 1.0 matches everything.
4929 0.0 makes pixels either fully transparent, or not transparent at all.
4931 Higher values result in semi-transparent pixels, with a higher transparency
4932 the more similar the pixels color is to the key color.
4935 Signals that the color passed is already in YUV instead of RGB.
4937 Litteral colors like "green" or "red" don't make sense with this enabled anymore.
4938 This can be used to pass exact YUV values as hexadecimal numbers.
4941 @subsection Examples
4945 Make every green pixel in the input image transparent:
4947 ffmpeg -i input.png -vf chromakey=green out.png
4951 Overlay a greenscreen-video on top of a static black background.
4953 ffmpeg -f lavfi -i color=c=black:s=1280x720 -i video.mp4 -shortest -filter_complex "[1:v]chromakey=0x70de77:0.1:0.2[ckout];[0:v][ckout]overlay[out]" -map "[out]" output.mkv
4959 Display CIE color diagram with pixels overlaid onto it.
4961 The filter accepts the following options:
4976 @item uhdtv, rec2020
4989 Set what gamuts to draw.
4991 See @code{system} option for available values.
4994 Set ciescope size, by default set to 512.
4997 Set intensity used to map input pixel values to CIE diagram.
5000 Set contrast used to draw tongue colors that are out of active color system gamut.
5003 Correct gamma displayed on scope, by default enabled.
5006 Show white point on CIE diagram, by default disabled.
5009 Set input gamma. Used only with XYZ input color space.
5014 Visualize information exported by some codecs.
5016 Some codecs can export information through frames using side-data or other
5017 means. For example, some MPEG based codecs export motion vectors through the
5018 @var{export_mvs} flag in the codec @option{flags2} option.
5020 The filter accepts the following option:
5024 Set motion vectors to visualize.
5026 Available flags for @var{mv} are:
5030 forward predicted MVs of P-frames
5032 forward predicted MVs of B-frames
5034 backward predicted MVs of B-frames
5038 Display quantization parameters using the chroma planes.
5041 Set motion vectors type to visualize. Includes MVs from all frames unless specified by @var{frame_type} option.
5043 Available flags for @var{mv_type} are:
5047 forward predicted MVs
5049 backward predicted MVs
5052 @item frame_type, ft
5053 Set frame type to visualize motion vectors of.
5055 Available flags for @var{frame_type} are:
5059 intra-coded frames (I-frames)
5061 predicted frames (P-frames)
5063 bi-directionally predicted frames (B-frames)
5067 @subsection Examples
5071 Visualize forward predicted MVs of all frames using @command{ffplay}:
5073 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv_type=fp
5077 Visualize multi-directionals MVs of P and B-Frames using @command{ffplay}:
5079 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv=pf+bf+bb
5083 @section colorbalance
5084 Modify intensity of primary colors (red, green and blue) of input frames.
5086 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
5087 regions for the red-cyan, green-magenta or blue-yellow balance.
5089 A positive adjustment value shifts the balance towards the primary color, a negative
5090 value towards the complementary color.
5092 The filter accepts the following options:
5098 Adjust red, green and blue shadows (darkest pixels).
5103 Adjust red, green and blue midtones (medium pixels).
5108 Adjust red, green and blue highlights (brightest pixels).
5110 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5113 @subsection Examples
5117 Add red color cast to shadows: