4 Encoders are configured elements in FFmpeg which allow the encoding of
7 When you configure your FFmpeg build, all the supported native encoders
8 are enabled by default. Encoders requiring an external library must be enabled
9 manually via the corresponding @code{--enable-lib} option. You can list all
10 available encoders using the configure option @code{--list-encoders}.
12 You can disable all the encoders with the configure option
13 @code{--disable-encoders} and selectively enable / disable single encoders
14 with the options @code{--enable-encoder=@var{ENCODER}} /
15 @code{--disable-encoder=@var{ENCODER}}.
17 The option @code{-codecs} of the ff* tools will display the list of
22 @chapter Audio Encoders
23 @c man begin AUDIO ENCODERS
25 A description of some of the currently available audio encoders
31 Advanced Audio Coding (AAC) encoder.
33 This encoder is an experimental FFmpeg-native AAC encoder. Currently only the
34 low complexity (AAC-LC) profile is supported. To use this encoder, you must set
35 @option{strict} option to @samp{experimental} or lower.
37 As this encoder is experimental, unexpected behavior may exist from time to
38 time. For a more stable AAC encoder, see @ref{libvo-aacenc}. However, be warned
39 that it has a worse quality reported by some users.
41 @c todo @ref{libaacplus}
42 See also @ref{libfdk-aac-enc,,libfdk_aac} and @ref{libfaac}.
48 Set bit rate in bits/s. Setting this automatically activates constant bit rate
52 Set quality for variable bit rate (VBR) mode. This option is valid only using
53 the @command{ffmpeg} command-line tool. For library interface users, use
54 @option{global_quality}.
57 Set stereo encoding mode. Possible values:
61 Automatically selected by the encoder.
64 Disable middle/side encoding. This is the default.
67 Force middle/side encoding.
71 Set AAC encoder coding method. Possible values:
77 This method is a simplified reimplementation of the method used in FAAC, which
78 sets thresholds proportional to the band energies, and then decreases all the
79 thresholds with quantizer steps to find the appropriate quantization with
80 distortion below threshold band by band.
82 The quality of this method is comparable to the two loop searching method
83 descibed below, but somewhat a little better and slower.
86 Average noise to mask ratio (ANMR) trellis-based solution.
88 This has a theoretic best quality out of all the coding methods, but at the
89 cost of the slowest speed.
92 Two loop searching (TLS) method.
94 This method first sets quantizers depending on band thresholds and then tries
95 to find an optimal combination by adding or subtracting a specific value from
96 all quantizers and adjusting some individual quantizer a little.
98 This method produces similar quality with the FAAC method and is the default.
101 Constant quantizer method.
103 This method sets a constant quantizer for all bands. This is the fastest of all
104 the methods, yet produces the worst quality.
110 @section ac3 and ac3_fixed
114 These encoders implement part of ATSC A/52:2010 and ETSI TS 102 366, as well as
115 the undocumented RealAudio 3 (a.k.a. dnet).
117 The @var{ac3} encoder uses floating-point math, while the @var{ac3_fixed}
118 encoder only uses fixed-point integer math. This does not mean that one is
119 always faster, just that one or the other may be better suited to a
120 particular system. The floating-point encoder will generally produce better
121 quality audio for a given bitrate. The @var{ac3_fixed} encoder is not the
122 default codec for any of the output formats, so it must be specified explicitly
123 using the option @code{-acodec ac3_fixed} in order to use it.
125 @subsection AC-3 Metadata
127 The AC-3 metadata options are used to set parameters that describe the audio,
128 but in most cases do not affect the audio encoding itself. Some of the options
129 do directly affect or influence the decoding and playback of the resulting
130 bitstream, while others are just for informational purposes. A few of the
131 options will add bits to the output stream that could otherwise be used for
132 audio data, and will thus affect the quality of the output. Those will be
133 indicated accordingly with a note in the option list below.
135 These parameters are described in detail in several publicly-available
138 @item @uref{http://www.atsc.org/cms/standards/a_52-2010.pdf,A/52:2010 - Digital Audio Compression (AC-3) (E-AC-3) Standard}
139 @item @uref{http://www.atsc.org/cms/standards/a_54a_with_corr_1.pdf,A/54 - Guide to the Use of the ATSC Digital Television Standard}
140 @item @uref{http://www.dolby.com/uploadedFiles/zz-_Shared_Assets/English_PDFs/Professional/18_Metadata.Guide.pdf,Dolby Metadata Guide}
141 @item @uref{http://www.dolby.com/uploadedFiles/zz-_Shared_Assets/English_PDFs/Professional/46_DDEncodingGuidelines.pdf,Dolby Digital Professional Encoding Guidelines}
144 @subsubsection Metadata Control Options
148 @item -per_frame_metadata @var{boolean}
149 Allow Per-Frame Metadata. Specifies if the encoder should check for changing
150 metadata for each frame.
153 The metadata values set at initialization will be used for every frame in the
156 Metadata values can be changed before encoding each frame.
161 @subsubsection Downmix Levels
165 @item -center_mixlev @var{level}
166 Center Mix Level. The amount of gain the decoder should apply to the center
167 channel when downmixing to stereo. This field will only be written to the
168 bitstream if a center channel is present. The value is specified as a scale
169 factor. There are 3 valid values:
174 Apply -4.5dB gain (default)
179 @item -surround_mixlev @var{level}
180 Surround Mix Level. The amount of gain the decoder should apply to the surround
181 channel(s) when downmixing to stereo. This field will only be written to the
182 bitstream if one or more surround channels are present. The value is specified
183 as a scale factor. There are 3 valid values:
188 Apply -6dB gain (default)
190 Silence Surround Channel(s)
195 @subsubsection Audio Production Information
196 Audio Production Information is optional information describing the mixing
197 environment. Either none or both of the fields are written to the bitstream.
201 @item -mixing_level @var{number}
202 Mixing Level. Specifies peak sound pressure level (SPL) in the production
203 environment when the mix was mastered. Valid values are 80 to 111, or -1 for
204 unknown or not indicated. The default value is -1, but that value cannot be
205 used if the Audio Production Information is written to the bitstream. Therefore,
206 if the @code{room_type} option is not the default value, the @code{mixing_level}
207 option must not be -1.
209 @item -room_type @var{type}
210 Room Type. Describes the equalization used during the final mixing session at
211 the studio or on the dubbing stage. A large room is a dubbing stage with the
212 industry standard X-curve equalization; a small room has flat equalization.
213 This field will not be written to the bitstream if both the @code{mixing_level}
214 option and the @code{room_type} option have the default values.
218 Not Indicated (default)
229 @subsubsection Other Metadata Options
233 @item -copyright @var{boolean}
234 Copyright Indicator. Specifies whether a copyright exists for this audio.
238 No Copyright Exists (default)
244 @item -dialnorm @var{value}
245 Dialogue Normalization. Indicates how far the average dialogue level of the
246 program is below digital 100% full scale (0 dBFS). This parameter determines a
247 level shift during audio reproduction that sets the average volume of the
248 dialogue to a preset level. The goal is to match volume level between program
249 sources. A value of -31dB will result in no volume level change, relative to
250 the source volume, during audio reproduction. Valid values are whole numbers in
251 the range -31 to -1, with -31 being the default.
253 @item -dsur_mode @var{mode}
254 Dolby Surround Mode. Specifies whether the stereo signal uses Dolby Surround
255 (Pro Logic). This field will only be written to the bitstream if the audio
256 stream is stereo. Using this option does @b{NOT} mean the encoder will actually
257 apply Dolby Surround processing.
261 Not Indicated (default)
264 Not Dolby Surround Encoded
267 Dolby Surround Encoded
270 @item -original @var{boolean}
271 Original Bit Stream Indicator. Specifies whether this audio is from the
272 original source and not a copy.
279 Original Source (default)
284 @subsection Extended Bitstream Information
285 The extended bitstream options are part of the Alternate Bit Stream Syntax as
286 specified in Annex D of the A/52:2010 standard. It is grouped into 2 parts.
287 If any one parameter in a group is specified, all values in that group will be
288 written to the bitstream. Default values are used for those that are written
289 but have not been specified. If the mixing levels are written, the decoder
290 will use these values instead of the ones specified in the @code{center_mixlev}
291 and @code{surround_mixlev} options if it supports the Alternate Bit Stream
294 @subsubsection Extended Bitstream Information - Part 1
298 @item -dmix_mode @var{mode}
299 Preferred Stereo Downmix Mode. Allows the user to select either Lt/Rt
300 (Dolby Surround) or Lo/Ro (normal stereo) as the preferred stereo downmix mode.
304 Not Indicated (default)
307 Lt/Rt Downmix Preferred
310 Lo/Ro Downmix Preferred
313 @item -ltrt_cmixlev @var{level}
314 Lt/Rt Center Mix Level. The amount of gain the decoder should apply to the
315 center channel when downmixing to stereo in Lt/Rt mode.
328 Apply -4.5dB gain (default)
332 Silence Center Channel
335 @item -ltrt_surmixlev @var{level}
336 Lt/Rt Surround Mix Level. The amount of gain the decoder should apply to the
337 surround channel(s) when downmixing to stereo in Lt/Rt mode.
346 Apply -6.0dB gain (default)
348 Silence Surround Channel(s)
351 @item -loro_cmixlev @var{level}
352 Lo/Ro Center Mix Level. The amount of gain the decoder should apply to the
353 center channel when downmixing to stereo in Lo/Ro mode.
366 Apply -4.5dB gain (default)
370 Silence Center Channel
373 @item -loro_surmixlev @var{level}
374 Lo/Ro Surround Mix Level. The amount of gain the decoder should apply to the
375 surround channel(s) when downmixing to stereo in Lo/Ro mode.
384 Apply -6.0dB gain (default)
386 Silence Surround Channel(s)
391 @subsubsection Extended Bitstream Information - Part 2
395 @item -dsurex_mode @var{mode}
396 Dolby Surround EX Mode. Indicates whether the stream uses Dolby Surround EX
397 (7.1 matrixed to 5.1). Using this option does @b{NOT} mean the encoder will actually
398 apply Dolby Surround EX processing.
402 Not Indicated (default)
405 Dolby Surround EX Off
411 @item -dheadphone_mode @var{mode}
412 Dolby Headphone Mode. Indicates whether the stream uses Dolby Headphone
413 encoding (multi-channel matrixed to 2.0 for use with headphones). Using this
414 option does @b{NOT} mean the encoder will actually apply Dolby Headphone
419 Not Indicated (default)
428 @item -ad_conv_type @var{type}
429 A/D Converter Type. Indicates whether the audio has passed through HDCD A/D
434 Standard A/D Converter (default)
442 @subsection Other AC-3 Encoding Options
446 @item -stereo_rematrixing @var{boolean}
447 Stereo Rematrixing. Enables/Disables use of rematrixing for stereo input. This
448 is an optional AC-3 feature that increases quality by selectively encoding
449 the left/right channels as mid/side. This option is enabled by default, and it
450 is highly recommended that it be left as enabled except for testing purposes.
454 @subsection Floating-Point-Only AC-3 Encoding Options
456 These options are only valid for the floating-point encoder and do not exist
457 for the fixed-point encoder due to the corresponding features not being
458 implemented in fixed-point.
462 @item -channel_coupling @var{boolean}
463 Enables/Disables use of channel coupling, which is an optional AC-3 feature
464 that increases quality by combining high frequency information from multiple
465 channels into a single channel. The per-channel high frequency information is
466 sent with less accuracy in both the frequency and time domains. This allows
467 more bits to be used for lower frequencies while preserving enough information
468 to reconstruct the high frequencies. This option is enabled by default for the
469 floating-point encoder and should generally be left as enabled except for
470 testing purposes or to increase encoding speed.
474 Selected by Encoder (default)
477 Disable Channel Coupling
480 Enable Channel Coupling
483 @item -cpl_start_band @var{number}
484 Coupling Start Band. Sets the channel coupling start band, from 1 to 15. If a
485 value higher than the bandwidth is used, it will be reduced to 1 less than the
486 coupling end band. If @var{auto} is used, the start band will be determined by
487 the encoder based on the bit rate, sample rate, and channel layout. This option
488 has no effect if channel coupling is disabled.
492 Selected by Encoder (default)
500 libfaac AAC (Advanced Audio Coding) encoder wrapper.
502 Requires the presence of the libfaac headers and library during
503 configuration. You need to explicitly configure the build with
504 @code{--enable-libfaac --enable-nonfree}.
506 This encoder is considered to be of higher quality with respect to the
507 @ref{aacenc,,the native experimental FFmpeg AAC encoder}.
509 For more information see the libfaac project at
510 @url{http://www.audiocoding.com/faac.html/}.
514 The following shared FFmpeg codec options are recognized.
516 The following options are supported by the libfaac wrapper. The
517 @command{faac}-equivalent of the options are listed in parentheses.
521 Set bit rate in bits/s for ABR (Average Bit Rate) mode. If the bit rate
522 is not explicitly specified, it is automatically set to a suitable
523 value depending on the selected profile. @command{faac} bitrate is
524 expressed in kilobits/s.
526 Note that libfaac does not support CBR (Constant Bit Rate) but only
527 ABR (Average Bit Rate).
529 If VBR mode is enabled this option is ignored.
532 Set audio sampling rate (in Hz).
535 Set the number of audio channels.
537 @item cutoff (@emph{-C})
538 Set cutoff frequency. If not specified (or explicitly set to 0) it
539 will use a value automatically computed by the library. Default value
545 The following profiles are recognized:
551 Low Complexity AAC (LC)
554 Scalable Sample Rate (SSR)
557 Long Term Prediction (LTP)
560 If not specified it is set to @samp{aac_low}.
563 Set constant quality VBR (Variable Bit Rate) mode.
566 Set quality in VBR mode as an integer number of lambda units.
568 Only relevant when VBR mode is enabled with @code{flags +qscale}. The
569 value is converted to QP units by dividing it by @code{FF_QP2LAMBDA},
570 and used to set the quality value used by libfaac. A reasonable range
571 for the option value in QP units is [10-500], the higher the value the
575 Enable VBR mode when set to a non-negative value, and set constant
576 quality value as a double floating point value in QP units.
578 The value sets the quality value used by libfaac. A reasonable range
579 for the option value is [10-500], the higher the value the higher the
582 This option is valid only using the @command{ffmpeg} command-line
583 tool. For library interface users, use @option{global_quality}.
590 Use @command{ffmpeg} to convert an audio file to ABR 128 kbps AAC in an M4A (MP4)
593 ffmpeg -i input.wav -codec:a libfaac -b:a 128k -output.m4a
597 Use @command{ffmpeg} to convert an audio file to VBR AAC, using the
600 ffmpeg -i input.wav -c:a libfaac -profile:a aac_ltp -q:a 100 output.m4a
604 @anchor{libfdk-aac-enc}
607 libfdk-aac AAC (Advanced Audio Coding) encoder wrapper.
609 The libfdk-aac library is based on the Fraunhofer FDK AAC code from
612 Requires the presence of the libfdk-aac headers and library during
613 configuration. You need to explicitly configure the build with
614 @code{--enable-libfdk-aac}. The library is also incompatible with GPL,
615 so if you allow the use of GPL, you should configure with
616 @code{--enable-gpl --enable-nonfree --enable-libfdk-aac}.
618 This encoder is considered to be of higher quality with respect to
619 both @ref{aacenc,,the native experimental FFmpeg AAC encoder} and
622 For more information see the fdk-aac project at
623 @url{http://sourceforge.net/p/opencore-amr/fdk-aac/}.
627 The following options are mapped on the shared FFmpeg codec options.
631 Set bit rate in bits/s. If the bitrate is not explicitly specified, it
632 is automatically set to a suitable value depending on the selected
635 In case VBR mode is enabled the option is ignored.
638 Set audio sampling rate (in Hz).
641 Set the number of audio channels.
644 Enable fixed quality, VBR (Variable Bit Rate) mode.
645 Note that VBR is implicitely enabled when the @option{vbr} value is
649 Set cutoff frequency. If not specified (or explicitly set to 0) it
650 will use a value automatically computed by the library. Default value
656 The following profiles are recognized:
659 Low Complexity AAC (LC)
662 High Efficiency AAC (HE-AAC)
665 High Efficiency AAC version 2 (HE-AACv2)
671 Enhanced Low Delay AAC (ELD)
674 If not specified it is set to @samp{aac_low}.
677 The following are private options of the libfdk_aac encoder.
681 Enable afterburner feature if set to 1, disabled if set to 0. This
682 improves the quality but also the required processing power.
687 Enable SBR (Spectral Band Replication) for ELD if set to 1, disabled
693 Set SBR/PS signaling style.
695 It can assume one of the following values:
698 choose signaling implicitly (explicit hierarchical by default,
699 implicit if global header is disabled)
702 implicit backwards compatible signaling
705 explicit SBR, implicit PS signaling
707 @item explicit_hierarchical
708 explicit hierarchical signaling
711 Default value is @samp{default}.
714 Output LATM/LOAS encapsulated data if set to 1, disabled if set to 0.
719 Set StreamMuxConfig and PCE repetition period (in frames) for sending
720 in-band configuration buffers within LATM/LOAS transport layer.
722 Must be a 16-bits non-negative integer.
727 Set VBR mode, from 1 to 5. 1 is lowest quality (though still pretty
728 good) and 5 is highest quality. A value of 0 will disable VBR mode,
729 and CBR (Constant Bit Rate) mode is enabled.
731 Currently only the @samp{aac_low} profile supports VBR mode.
733 VBR modes 1-5 correspond to roughly the following average bit rate:
745 about 80-96 kbps/channel
755 Use @command{ffmpeg} to convert an audio file to VBR AAC in an M4A (MP4)
758 ffmpeg -i input.wav -codec:a libfdk_aac -vbr 3 output.m4a
762 Use @command{ffmpeg} to convert an audio file to CBR 64k kbps AAC, using the
763 High-Efficiency AAC profile:
765 ffmpeg -i input.wav -c:a libfdk_aac -profile:a aac_he -b:a 64k output.m4a
772 LAME (Lame Ain't an MP3 Encoder) MP3 encoder wrapper.
774 Requires the presence of the libmp3lame headers and library during
775 configuration. You need to explicitly configure the build with
776 @code{--enable-libmp3lame}.
778 See @ref{libshine} for a fixed-point MP3 encoder, although with a
783 The following options are supported by the libmp3lame wrapper. The
784 @command{lame}-equivalent of the options are listed in parentheses.
788 Set bitrate expressed in bits/s for CBR or ABR. LAME @code{bitrate} is
789 expressed in kilobits/s.
792 Set constant quality setting for VBR. This option is valid only
793 using the @command{ffmpeg} command-line tool. For library interface
794 users, use @option{global_quality}.
796 @item compression_level (@emph{-q})
797 Set algorithm quality. Valid arguments are integers in the 0-9 range,
798 with 0 meaning highest quality but slowest, and 9 meaning fastest
799 while producing the worst quality.
802 Enable use of bit reservoir when set to 1. Default value is 1. LAME
803 has this enabled by default, but can be overriden by use
804 @option{--nores} option.
806 @item joint_stereo (@emph{-m j})
807 Enable the encoder to use (on a frame by frame basis) either L/R
808 stereo or mid/side stereo. Default value is 1.
810 @item abr (@emph{--abr})
811 Enable the encoder to use ABR when set to 1. The @command{lame}
812 @option{--abr} sets the target bitrate, while this options only
813 tells FFmpeg to use ABR still relies on @option{b} to set bitrate.
817 @section libopencore-amrnb
819 OpenCORE Adaptive Multi-Rate Narrowband encoder.
821 Requires the presence of the libopencore-amrnb headers and library during
822 configuration. You need to explicitly configure the build with
823 @code{--enable-libopencore-amrnb --enable-version3}.
825 This is a mono-only encoder. Officially it only supports 8000Hz sample rate,
826 but you can override it by setting @option{strict} to @samp{unofficial} or
834 Set bitrate in bits per second. Only the following bitrates are supported,
835 otherwise libavcodec will round to the nearest valid bitrate.
849 Allow discontinuous transmission (generate comfort noise) when set to 1. The
850 default value is 0 (disabled).
857 Shine Fixed-Point MP3 encoder wrapper.
859 Shine is a fixed-point MP3 encoder. It has a far better performance on
860 platforms without an FPU, e.g. armel CPUs, and some phones and tablets.
861 However, as it is more targeted on performance than quality, it is not on par
862 with LAME and other production-grade encoders quality-wise. Also, according to
863 the project's homepage, this encoder may not be free of bugs as the code was
864 written a long time ago and the project was dead for at least 5 years.
866 This encoder only supports stereo and mono input. This is also CBR-only.
868 The original project (last updated in early 2007) is at
869 @url{http://sourceforge.net/projects/libshine-fxp/}. We only support the
870 updated fork by the Savonet/Liquidsoap project at @url{https://github.com/savonet/shine}.
872 Requires the presence of the libshine headers and library during
873 configuration. You need to explicitly configure the build with
874 @code{--enable-libshine}.
876 See also @ref{libmp3lame}.
880 The following options are supported by the libshine wrapper. The
881 @command{shineenc}-equivalent of the options are listed in parentheses.
885 Set bitrate expressed in bits/s for CBR. @command{shineenc} @option{-b} option
886 is expressed in kilobits/s.
892 TwoLAME MP2 encoder wrapper.
894 Requires the presence of the libtwolame headers and library during
895 configuration. You need to explicitly configure the build with
896 @code{--enable-libtwolame}.
900 The following options are supported by the libtwolame wrapper. The
901 @command{twolame}-equivalent options follow the FFmpeg ones and are in
906 Set bitrate expressed in bits/s for CBR. @command{twolame} @option{b}
907 option is expressed in kilobits/s. Default value is 128k.
910 Set quality for experimental VBR support. Maximum value range is
911 from -50 to 50, useful range is from -10 to 10. The higher the
912 value, the better the quality. This option is valid only using the
913 @command{ffmpeg} command-line tool. For library interface users,
914 use @option{global_quality}.
916 @item mode (@emph{--mode})
917 Set the mode of the resulting audio. Possible values:
921 Choose mode automatically based on the input. This is the default.
932 @item psymodel (@emph{--psyc-mode})
933 Set psychoacoustic model to use in encoding. The argument must be
934 an integer between -1 and 4, inclusive. The higher the value, the
935 better the quality. The default value is 3.
937 @item energy_levels (@emph{--energy})
938 Enable energy levels extensions when set to 1. The default value is
941 @item error_protection (@emph{--protect})
942 Enable CRC error protection when set to 1. The default value is 0
945 @item copyright (@emph{--copyright})
946 Set MPEG audio copyright flag when set to 1. The default value is 0
949 @item original (@emph{--original})
950 Set MPEG audio original flag when set to 1. The default value is 0
955 @anchor{libvo-aacenc}
956 @section libvo-aacenc
958 VisualOn AAC encoder.
960 Requires the presence of the libvo-aacenc headers and library during
961 configuration. You need to explicitly configure the build with
962 @code{--enable-libvo-aacenc --enable-version3}.
964 This encoder is considered to be worse than the
965 @ref{aacenc,,native experimental FFmpeg AAC encoder}, according to
970 The VisualOn AAC encoder only support encoding AAC-LC and up to 2
971 channels. It is also CBR-only.
976 Set bit rate in bits/s.
980 @section libvo-amrwbenc
982 VisualOn Adaptive Multi-Rate Wideband encoder.
984 Requires the presence of the libvo-amrwbenc headers and library during
985 configuration. You need to explicitly configure the build with
986 @code{--enable-libvo-amrwbenc --enable-version3}.
988 This is a mono-only encoder. Officially it only supports 16000Hz sample
989 rate, but you can override it by setting @option{strict} to
990 @samp{unofficial} or lower.
997 Set bitrate in bits/s. Only the following bitrates are supported, otherwise
998 libavcodec will round to the nearest valid bitrate.
1013 Allow discontinuous transmission (generate comfort noise) when set to 1. The
1014 default value is 0 (disabled).
1020 libopus Opus Interactive Audio Codec encoder wrapper.
1022 Requires the presence of the libopus headers and library during
1023 configuration. You need to explicitly configure the build with
1024 @code{--enable-libopus}.
1026 @subsection Option Mapping
1028 Most libopus options are modeled after the @command{opusenc} utility from
1029 opus-tools. The following is an option mapping chart describing options
1030 supported by the libopus wrapper, and their @command{opusenc}-equivalent
1035 @item b (@emph{bitrate})
1036 Set the bit rate in bits/s. FFmpeg's @option{b} option is
1037 expressed in bits/s, while @command{opusenc}'s @option{bitrate} in
1040 @item vbr (@emph{vbr}, @emph{hard-cbr}, and @emph{cvbr})
1041 Set VBR mode. The FFmpeg @option{vbr} option has the following
1042 valid arguments, with the their @command{opusenc} equivalent options
1046 @item off (@emph{hard-cbr})
1047 Use constant bit rate encoding.
1049 @item on (@emph{vbr})
1050 Use variable bit rate encoding (the default).
1052 @item constrained (@emph{cvbr})
1053 Use constrained variable bit rate encoding.
1056 @item compression_level (@emph{comp})
1057 Set encoding algorithm complexity. Valid options are integers in
1058 the 0-10 range. 0 gives the fastest encodes but lower quality, while 10
1059 gives the highest quality but slowest encoding. The default is 10.
1061 @item frame_duration (@emph{framesize})
1062 Set maximum frame size, or duration of a frame in milliseconds. The
1063 argument must be exactly the following: 2.5, 5, 10, 20, 40, 60. Smaller
1064 frame sizes achieve lower latency but less quality at a given bitrate.
1065 Sizes greater than 20ms are only interesting at fairly low bitrates.
1066 The default is 20ms.
1068 @item packet_loss (@emph{expect-loss})
1069 Set expected packet loss percentage. The default is 0.
1071 @item application (N.A.)
1072 Set intended application type. Valid options are listed below:
1076 Favor improved speech intelligibility.
1078 Favor faithfulness to the input (the default).
1080 Restrict to only the lowest delay modes.
1084 Set cutoff bandwidth in Hz. The argument must be exactly one of the
1085 following: 4000, 6000, 8000, 12000, or 20000, corresponding to
1086 narrowband, mediumband, wideband, super wideband, and fullband
1087 respectively. The default is 0 (cutoff disabled).
1093 libvorbis encoder wrapper.
1095 Requires the presence of the libvorbisenc headers and library during
1096 configuration. You need to explicitly configure the build with
1097 @code{--enable-libvorbis}.
1101 The following options are supported by the libvorbis wrapper. The
1102 @command{oggenc}-equivalent of the options are listed in parentheses.
1104 To get a more accurate and extensive documentation of the libvorbis
1105 options, consult the libvorbisenc's and @command{oggenc}'s documentations.
1106 See @url{http://xiph.org/vorbis/},
1107 @url{http://wiki.xiph.org/Vorbis-tools}, and oggenc(1).
1111 Set bitrate expressed in bits/s for ABR. @command{oggenc} @option{-b} is
1112 expressed in kilobits/s.
1115 Set constant quality setting for VBR. The value should be a float
1116 number in the range of -1.0 to 10.0. The higher the value, the better
1117 the quality. The default value is @samp{3.0}.
1119 This option is valid only using the @command{ffmpeg} command-line tool.
1120 For library interface users, use @option{global_quality}.
1122 @item cutoff (@emph{--advanced-encode-option lowpass_frequency=N})
1123 Set cutoff bandwidth in Hz, a value of 0 disables cutoff. @command{oggenc}'s
1124 related option is expressed in kHz. The default value is @samp{0} (cutoff
1127 @item minrate (@emph{-m})
1128 Set minimum bitrate expressed in bits/s. @command{oggenc} @option{-m} is
1129 expressed in kilobits/s.
1131 @item maxrate (@emph{-M})
1132 Set maximum bitrate expressed in bits/s. @command{oggenc} @option{-M} is
1133 expressed in kilobits/s. This only has effect on ABR mode.
1135 @item iblock (@emph{--advanced-encode-option impulse_noisetune=N})
1136 Set noise floor bias for impulse blocks. The value is a float number from
1137 -15.0 to 0.0. A negative bias instructs the encoder to pay special attention
1138 to the crispness of transients in the encoded audio. The tradeoff for better
1139 transient response is a higher bitrate.
1145 A wrapper providing WavPack encoding through libwavpack.
1147 Only lossless mode using 32-bit integer samples is supported currently.
1148 The @option{compression_level} option can be used to control speed vs.
1149 compression tradeoff, with the values mapped to libwavpack as follows:
1154 Fast mode - corresponding to the wavpack @option{-f} option.
1157 Normal (default) settings.
1160 High quality - corresponding to the wavpack @option{-h} option.
1163 Very high quality - corresponding to the wavpack @option{-hh} option.
1166 Same as 3, but with extra processing enabled - corresponding to the wavpack
1167 @option{-x} option. I.e. 4 is the same as @option{-x2} and 8 is the same as
1172 @c man end AUDIO ENCODERS
1174 @chapter Video Encoders
1175 @c man begin VIDEO ENCODERS
1177 A description of some of the currently available video encoders
1182 libtheora Theora encoder wrapper.
1184 Requires the presence of the libtheora headers and library during
1185 configuration. You need to explicitly configure the build with
1186 @code{--enable-libtheora}.
1188 For more informations about the libtheora project see
1189 @url{http://www.theora.org/}.
1193 The following global options are mapped to internal libtheora options
1194 which affect the quality and the bitrate of the encoded stream.
1198 Set the video bitrate in bit/s for CBR (Constant Bit Rate) mode. In
1199 case VBR (Variable Bit Rate) mode is enabled this option is ignored.
1202 Used to enable constant quality mode (VBR) encoding through the
1203 @option{qscale} flag, and to enable the @code{pass1} and @code{pass2}
1209 @item global_quality
1210 Set the global quality as an integer in lambda units.
1212 Only relevant when VBR mode is enabled with @code{flags +qscale}. The
1213 value is converted to QP units by dividing it by @code{FF_QP2LAMBDA},
1214 clipped in the [0 - 10] range, and then multiplied by 6.3 to get a
1215 value in the native libtheora range [0-63]. A higher value corresponds
1216 to a higher quality.
1219 Enable VBR mode when set to a non-negative value, and set constant
1220 quality value as a double floating point value in QP units.
1222 The value is clipped in the [0-10] range, and then multiplied by 6.3
1223 to get a value in the native libtheora range [0-63].
1225 This option is valid only using the @command{ffmpeg} command-line
1226 tool. For library interface users, use @option{global_quality}.
1229 @subsection Examples
1233 Set maximum constant quality (VBR) encoding with @command{ffmpeg}:
1235 ffmpeg -i INPUT -codec:v libtheora -q:v 10 OUTPUT.ogg
1239 Use @command{ffmpeg} to convert a CBR 1000 kbps Theora video stream:
1241 ffmpeg -i INPUT -codec:v libtheora -b:v 1000k OUTPUT.ogg
1247 VP8 format supported through libvpx.
1249 Requires the presence of the libvpx headers and library during configuration.
1250 You need to explicitly configure the build with @code{--enable-libvpx}.
1254 Mapping from FFmpeg to libvpx options with conversion notes in parentheses.
1281 @code{(bufsize * 1000 / vb)}
1284 @code{(bufsize * 1000 / vb * 5 / 6)}
1286 @item rc_init_occupancy, vb
1288 @code{(rc_init_occupancy * 1000 / vb)}
1290 @item rc_buffer_aggressivity
1293 @item skip_threshold
1297 rc_2pass_vbr_bias_pct
1300 rc_2pass_vbr_maxsection_pct
1301 @code{(maxrate * 100 / vb)}
1304 rc_2pass_vbr_minsection_pct
1305 @code{(minrate * 100 / vb)}
1307 @item minrate, maxrate, vb
1309 @code{(minrate == maxrate == vb)}
1312 @code{VPX_CQ}, @code{VP8E_SET_CQ_LEVEL}
1317 @code{VPX_DL_BEST_QUALITY}
1319 @code{VPX_DL_GOOD_QUALITY}
1320 @item @var{realtime}
1321 @code{VPX_DL_REALTIME}
1325 @code{VP8E_SET_CPUUSED}
1328 @code{VP8E_SET_NOISE_SENSITIVITY}
1331 @code{VP8E_SET_STATIC_THRESHOLD}
1334 @code{VP8E_SET_TOKEN_PARTITIONS}
1336 @item max-intra-rate
1337 @code{VP8E_SET_MAX_INTRA_BITRATE_PCT}
1339 @item force_key_frames
1340 @code{VPX_EFLAG_FORCE_KF}
1342 @item Alternate reference frame related
1344 @item vp8flags altref
1345 @code{VP8E_SET_ENABLEAUTOALTREF}
1346 @item @var{arnr_max_frames}
1347 @code{VP8E_SET_ARNR_MAXFRAMES}
1348 @item @var{arnr_type}
1349 @code{VP8E_SET_ARNR_TYPE}
1350 @item @var{arnr_strength}
1351 @code{VP8E_SET_ARNR_STRENGTH}
1352 @item @var{rc_lookahead}
1356 @item vp8flags error_resilient
1361 For more information about libvpx see:
1362 @url{http://www.webmproject.org/}
1366 x264 H.264/MPEG-4 AVC encoder wrapper.
1368 This encoder requires the presence of the libx264 headers and library
1369 during configuration. You need to explicitly configure the build with
1370 @code{--enable-libx264}.
1372 libx264 supports an impressive number of features, including 8x8 and
1373 4x4 adaptive spatial transform, adaptive B-frame placement, CAVLC/CABAC
1374 entropy coding, interlacing (MBAFF), lossless mode, psy optimizations
1375 for detail retention (adaptive quantization, psy-RD, psy-trellis).
1377 Many libx264 encoder options are mapped to FFmpeg global codec
1378 options, while unique encoder options are provided through private
1379 options. Additionally the @option{x264opts} and @option{x264-params}
1380 private options allows to pass a list of key=value tuples as accepted
1381 by the libx264 @code{x264_param_parse} function.
1383 The x264 project website is at
1384 @url{http://www.videolan.org/developers/x264.html}.
1388 The following options are supported by the libx264 wrapper. The
1389 @command{x264}-equivalent options or values are listed in parentheses
1392 To reduce the duplication of documentation, only the private options
1393 and some others requiring special attention are documented here. For
1394 the documentation of the undocumented generic options, see
1395 @ref{codec-options,,the Codec Options chapter}.
1397 To get a more accurate and extensive documentation of the libx264
1398 options, invoke the command @command{x264 --full-help} or consult
1399 the libx264 documentation.
1402 @item b (@emph{bitrate})
1403 Set bitrate in bits/s. Note that FFmpeg's @option{b} option is
1404 expressed in bits/s, while @command{x264}'s @option{bitrate} is in
1407 @item bf (@emph{bframes})
1409 @item g (@emph{keyint})
1411 @item qmax (@emph{qpmax})
1413 @item qmin (@emph{qpmin})
1415 @item qdiff (@emph{qpstep})
1417 @item qblur (@emph{qblur})
1419 @item qcomp (@emph{qcomp})
1421 @item refs (@emph{ref})
1423 @item sc_threshold (@emph{scenecut})
1425 @item trellis (@emph{trellis})
1427 @item nr (@emph{nr})
1429 @item me_range (@emph{merange})
1431 @item me_method (@emph{me})
1432 Set motion estimation method. Possible values in the decreasing order
1436 @item dia (@emph{dia})
1437 @item epzs (@emph{dia})
1438 Diamond search with radius 1 (fastest). @samp{epzs} is an alias for
1440 @item hex (@emph{hex})
1441 Hexagonal search with radius 2.
1442 @item umh (@emph{umh})
1443 Uneven multi-hexagon search.
1444 @item esa (@emph{esa})
1446 @item tesa (@emph{tesa})
1447 Hadamard exhaustive search (slowest).
1450 @item subq (@emph{subme})
1452 @item b_strategy (@emph{b-adapt})
1454 @item keyint_min (@emph{min-keyint})
1457 Set entropy encoder. Possible values:
1464 Enable CAVLC and disable CABAC. It generates the same effect as
1465 @command{x264}'s @option{--no-cabac} option.
1469 Set full pixel motion estimation comparation algorithm. Possible values:
1473 Enable chroma in motion estimation.
1476 Ignore chroma in motion estimation. It generates the same effect as
1477 @command{x264}'s @option{--no-chroma-me} option.
1480 @item threads (@emph{threads})
1483 Set multithreading technique. Possible values:
1487 Slice-based multithreading. It generates the same effect as
1488 @command{x264}'s @option{--sliced-threads} option.
1490 Frame-based multithreading.
1494 Set encoding flags. It can be used to disable closed GOP and enable
1495 open GOP by setting it to @code{-cgop}. The result is similar to
1496 the behavior of @command{x264}'s @option{--open-gop} option.
1498 @item rc_init_occupancy (@emph{vbv-init})
1500 @item preset (@emph{preset})
1501 Set the encoding preset.
1503 @item tune (@emph{tune})
1504 Set tuning of the encoding params.
1506 @item profile (@emph{profile})
1507 Set profile restrictions.
1510 Enable fast settings when encoding first pass, when set to 1. When set
1511 to 0, it has the same effect of @command{x264}'s
1512 @option{--slow-firstpass} option.
1514 @item crf (@emph{crf})
1515 Set the quality for constant quality mode.
1517 @item crf_max (@emph{crf-max})
1518 In CRF mode, prevents VBV from lowering quality beyond this point.
1520 @item qp (@emph{qp})
1521 Set constant quantization rate control method parameter.
1523 @item aq-mode (@emph{aq-mode})
1524 Set AQ method. Possible values:
1527 @item none (@emph{0})
1530 @item variance (@emph{1})
1531 Variance AQ (complexity mask).
1533 @item autovariance (@emph{2})
1534 Auto-variance AQ (experimental).
1537 @item aq-strength (@emph{aq-strength})
1538 Set AQ strength, reduce blocking and blurring in flat and textured areas.
1541 Use psychovisual optimizations when set to 1. When set to 0, it has the
1542 same effect as @command{x264}'s @option{--no-psy} option.
1544 @item psy-rd (@emph{psy-rd})
1545 Set strength of psychovisual optimization, in
1546 @var{psy-rd}:@var{psy-trellis} format.
1548 @item rc-lookahead (@emph{rc-lookahead})
1549 Set number of frames to look ahead for frametype and ratecontrol.
1552 Enable weighted prediction for B-frames when set to 1. When set to 0,
1553 it has the same effect as @command{x264}'s @option{--no-weightb} option.
1555 @item weightp (@emph{weightp})
1556 Set weighted prediction method for P-frames. Possible values:
1559 @item none (@emph{0})
1561 @item simple (@emph{1})
1562 Enable only weighted refs
1563 @item smart (@emph{2})
1564 Enable both weighted refs and duplicates
1567 @item ssim (@emph{ssim})
1568 Enable calculation and printing SSIM stats after the encoding.
1570 @item intra-refresh (@emph{intra-refresh})
1571 Enable the use of Periodic Intra Refresh instead of IDR frames when set
1574 @item bluray-compat (@emph{bluray-compat})
1575 Configure the encoder to be compatible with the bluray standard.
1576 It is a shorthand for setting "bluray-compat=1 force-cfr=1".
1578 @item b-bias (@emph{b-bias})
1579 Set the influence on how often B-frames are used.
1581 @item b-pyramid (@emph{b-pyramid})
1582 Set method for keeping of some B-frames as references. Possible values:
1585 @item none (@emph{none})
1587 @item strict (@emph{strict})
1588 Strictly hierarchical pyramid.
1589 @item normal (@emph{normal})
1590 Non-strict (not Blu-ray compatible).
1594 Enable the use of one reference per partition, as opposed to one
1595 reference per macroblock when set to 1. When set to 0, it has the
1596 same effect as @command{x264}'s @option{--no-mixed-refs} option.
1599 Enable adaptive spatial transform (high profile 8x8 transform)
1600 when set to 1. When set to 0, it has the same effect as
1601 @command{x264}'s @option{--no-8x8dct} option.
1604 Enable early SKIP detection on P-frames when set to 1. When set
1605 to 0, it has the same effect as @command{x264}'s
1606 @option{--no-fast-pskip} option.
1608 @item aud (@emph{aud})
1609 Enable use of access unit delimiters when set to 1.
1612 Enable use macroblock tree ratecontrol when set to 1. When set
1613 to 0, it has the same effect as @command{x264}'s
1614 @option{--no-mbtree} option.
1616 @item deblock (@emph{deblock})
1617 Set loop filter parameters, in @var{alpha}:@var{beta} form.
1619 @item cplxblur (@emph{cplxblur})
1620 Set fluctuations reduction in QP (before curve compression).
1622 @item partitions (@emph{partitions})
1623 Set partitions to consider as a comma-separated list of. Possible
1628 8x8 P-frame partition.
1630 4x4 P-frame partition.
1632 4x4 B-frame partition.
1634 8x8 I-frame partition.
1636 4x4 I-frame partition.
1637 (Enabling @samp{p4x4} requires @samp{p8x8} to be enabled. Enabling
1638 @samp{i8x8} requires adaptive spatial transform (@option{8x8dct}
1639 option) to be enabled.)
1640 @item none (@emph{none})
1641 Do not consider any partitions.
1642 @item all (@emph{all})
1643 Consider every partition.
1646 @item direct-pred (@emph{direct})
1647 Set direct MV prediction mode. Possible values:
1650 @item none (@emph{none})
1651 Disable MV prediction.
1652 @item spatial (@emph{spatial})
1653 Enable spatial predicting.
1654 @item temporal (@emph{temporal})
1655 Enable temporal predicting.
1656 @item auto (@emph{auto})
1657 Automatically decided.
1660 @item slice-max-size (@emph{slice-max-size})
1661 Set the limit of the size of each slice in bytes. If not specified
1662 but RTP payload size (@option{ps}) is specified, that is used.
1664 @item stats (@emph{stats})
1665 Set the file name for multi-pass stats.
1667 @item nal-hrd (@emph{nal-hrd})
1668 Set signal HRD information (requires @option{vbv-bufsize} to be set).
1672 @item none (@emph{none})
1673 Disable HRD information signaling.
1674 @item vbr (@emph{vbr})
1676 @item cbr (@emph{cbr})
1677 Constant bit rate (not allowed in MP4 container).
1680 @item x264opts (N.A.)
1681 Set any x264 option, see @command{x264 --fullhelp} for a list.
1683 Argument is a list of @var{key}=@var{value} couples separated by
1684 ":". In @var{filter} and @var{psy-rd} options that use ":" as a separator
1685 themselves, use "," instead. They accept it as well since long ago but this
1686 is kept undocumented for some reason.
1688 For example to specify libx264 encoding options with @command{ffmpeg}:
1690 ffmpeg -i foo.mpg -vcodec libx264 -x264opts keyint=123:min-keyint=20 -an out.mkv
1693 @item x264-params (N.A.)
1694 Override the x264 configuration using a :-separated list of key=value
1697 This option is functionally the same as the @option{x264opts}, but is
1698 duplicated for compability with the Libav fork.
1700 For example to specify libx264 encoding options with @command{ffmpeg}:
1702 ffmpeg -i INPUT -c:v libx264 -x264-params level=30:bframes=0:weightp=0:\
1703 cabac=0:ref=1:vbv-maxrate=768:vbv-bufsize=2000:analyse=all:me=umh:\
1704 no-fast-pskip=1:subq=6:8x8dct=0:trellis=0 OUTPUT
1708 Encoding ffpresets for common usages are provided so they can be used with the
1709 general presets system (e.g. passing the @option{pre} option).
1713 Xvid MPEG-4 Part 2 encoder wrapper.
1715 This encoder requires the presence of the libxvidcore headers and library
1716 during configuration. You need to explicitly configure the build with
1717 @code{--enable-libxvid --enable-gpl}.
1719 The native @code{mpeg4} encoder supports the MPEG-4 Part 2 format, so
1720 users can encode to this format without this library.
1724 The following options are supported by the libxvid wrapper. Some of
1725 the following options are listed but are not documented, and
1726 correspond to shared codec options. See @ref{codec-options,,the Codec
1727 Options chapter} for their documentation. The other shared options
1728 which are not listed have no effect for the libxvid encoder.
1750 Set specific encoding flags. Possible values:
1755 Use four motion vector by macroblock.
1758 Enable high quality AC prediction.
1761 Only encode grayscale.
1764 Enable the use of global motion compensation (GMC).
1767 Enable quarter-pixel motion compensation.
1773 Place global headers in extradata instead of every keyframe.
1780 Set motion estimation method. Possible values in decreasing order of
1781 speed and increasing order of quality:
1785 Use no motion estimation (default).
1790 Enable advanced diamond zonal search for 16x16 blocks and half-pixel
1791 refinement for 16x16 blocks. @samp{x1} and @samp{log} are aliases for
1795 Enable all of the things described above, plus advanced diamond zonal
1796 search for 8x8 blocks, half-pixel refinement for 8x8 blocks, and motion
1797 estimation on chroma planes.
1800 Enable all of the things described above, plus extended 16x16 and 8x8
1805 Set macroblock decision algorithm. Possible values in the increasing
1810 Use macroblock comparing function algorithm (default).
1813 Enable rate distortion-based half pixel and quarter pixel refinement for
1817 Enable all of the things described above, plus rate distortion-based
1818 half pixel and quarter pixel refinement for 8x8 blocks, and rate
1819 distortion-based search using square pattern.
1823 Enable lumi masking adaptive quantization when set to 1. Default is 0
1827 Enable variance adaptive quantization when set to 1. Default is 0
1830 When combined with @option{lumi_aq}, the resulting quality will not
1831 be better than any of the two specified individually. In other
1832 words, the resulting quality will be the worse one of the two
1836 Set structural similarity (SSIM) displaying method. Possible values:
1840 Disable displaying of SSIM information.
1843 Output average SSIM at the end of encoding to stdout. The format of
1844 showing the average SSIM is:
1850 For users who are not familiar with C, %f means a float number, or
1851 a decimal (e.g. 0.939232).
1854 Output both per-frame SSIM data during encoding and average SSIM at
1855 the end of encoding to stdout. The format of per-frame information
1859 SSIM: avg: %1.3f min: %1.3f max: %1.3f
1862 For users who are not familiar with C, %1.3f means a float number
1863 rounded to 3 digits after the dot (e.g. 0.932).
1868 Set SSIM accuracy. Valid options are integers within the range of
1869 0-4, while 0 gives the most accurate result and 4 computes the
1878 @subsection Private options
1881 @item dpi @var{integer}
1882 Set physical density of pixels, in dots per inch, unset by default
1883 @item dpm @var{integer}
1884 Set physical density of pixels, in dots per meter, unset by default
1889 Apple ProRes encoder.
1891 FFmpeg contains 2 ProRes encoders, the prores-aw and prores-ks encoder.
1892 The used encoder can be choosen with the @code{-vcodec} option.
1894 @subsection Private Options for prores-ks
1897 @item profile @var{integer}
1898 Select the ProRes profile to encode
1907 @item quant_mat @var{integer}
1908 Select quantization matrix.
1917 If set to @var{auto}, the matrix matching the profile will be picked.
1918 If not set, the matrix providing the highest quality, @var{default}, will be
1921 @item bits_per_mb @var{integer}
1922 How many bits to allot for coding one macroblock. Different profiles use
1923 between 200 and 2400 bits per macroblock, the maximum is 8000.
1925 @item mbs_per_slice @var{integer}
1926 Number of macroblocks in each slice (1-8); the default value (8)
1927 should be good in almost all situations.
1929 @item vendor @var{string}
1930 Override the 4-byte vendor ID.
1931 A custom vendor ID like @var{apl0} would claim the stream was produced by
1934 @item alpha_bits @var{integer}
1935 Specify number of bits for alpha component.
1936 Possible values are @var{0}, @var{8} and @var{16}.
1937 Use @var{0} to disable alpha plane coding.
1941 @subsection Speed considerations
1943 In the default mode of operation the encoder has to honor frame constraints
1944 (i.e. not produc frames with size bigger than requested) while still making
1945 output picture as good as possible.
1946 A frame containing a lot of small details is harder to compress and the encoder
1947 would spend more time searching for appropriate quantizers for each slice.
1949 Setting a higher @option{bits_per_mb} limit will improve the speed.
1951 For the fastest encoding speed set the @option{qscale} parameter (4 is the
1952 recommended value) and do not set a size constraint.
1954 @c man end VIDEO ENCODERS