4 Muxers are configured elements in FFmpeg which allow writing
5 multimedia streams to a particular type of file.
7 When you configure your FFmpeg build, all the supported muxers
8 are enabled by default. You can list all available muxers using the
9 configure option @code{--list-muxers}.
11 You can disable all the muxers with the configure option
12 @code{--disable-muxers} and selectively enable / disable single muxers
13 with the options @code{--enable-muxer=@var{MUXER}} /
14 @code{--disable-muxer=@var{MUXER}}.
16 The option @code{-formats} of the ff* tools will display the list of
19 A description of some of the currently available muxers follows.
24 Audio Interchange File Format muxer.
28 It accepts the following options:
32 Enable ID3v2 tags writing when set to 1. Default is 0 (disabled).
35 Select ID3v2 version to write. Currently only version 3 and 4 (aka.
36 ID3v2.3 and ID3v2.4) are supported. The default is version 4.
43 Advanced Systems Format muxer.
45 Note that Windows Media Audio (wma) and Windows Media Video (wmv) use this
50 It accepts the following options:
54 Set the muxer packet size. By tuning this setting you may reduce data
55 fragmentation or muxer overhead depending on your source. Default value is
56 3200, minimum is 100, maximum is 64k.
63 Chromaprint fingerprinter
65 This muxer feeds audio data to the Chromaprint library, which generates
66 a fingerprint for the provided audio data. It takes a single signed
67 native-endian 16-bit raw audio stream.
72 @item silence_threshold
73 Threshold for detecting silence, ranges from 0 to 32767. -1 for default
74 (required for use with the AcoustID service).
77 Algorithm index to fingerprint with.
80 Format to output the fingerprint as. Accepts the following options:
83 Binary raw fingerprint
86 Binary compressed fingerprint
89 Base64 compressed fingerprint
98 CRC (Cyclic Redundancy Check) testing format.
100 This muxer computes and prints the Adler-32 CRC of all the input audio
101 and video frames. By default audio frames are converted to signed
102 16-bit raw audio and video frames to raw video before computing the
105 The output of the muxer consists of a single line of the form:
106 CRC=0x@var{CRC}, where @var{CRC} is a hexadecimal number 0-padded to
107 8 digits containing the CRC for all the decoded input frames.
109 See also the @ref{framecrc} muxer.
113 For example to compute the CRC of the input, and store it in the file
116 ffmpeg -i INPUT -f crc out.crc
119 You can print the CRC to stdout with the command:
121 ffmpeg -i INPUT -f crc -
124 You can select the output format of each frame with @command{ffmpeg} by
125 specifying the audio and video codec and format. For example to
126 compute the CRC of the input audio converted to PCM unsigned 8-bit
127 and the input video converted to MPEG-2 video, use the command:
129 ffmpeg -i INPUT -c:a pcm_u8 -c:v mpeg2video -f crc -
134 Adobe Flash Video Format muxer.
136 This muxer accepts the following options:
140 @item flvflags @var{flags}
145 @item aac_seq_header_detect
146 Place AAC sequence header based on audio stream data.
148 @item no_sequence_end
149 Disable sequence end tag.
152 Disable metadata tag.
154 @item no_duration_filesize
155 Disable duration and filesize in metadata when they are equal to zero
156 at the end of stream. (Be used to non-seekable living stream).
158 @item add_keyframe_index
159 Used to facilitate seeking; particularly for HTTP pseudo streaming.
166 Per-packet CRC (Cyclic Redundancy Check) testing format.
168 This muxer computes and prints the Adler-32 CRC for each audio
169 and video packet. By default audio frames are converted to signed
170 16-bit raw audio and video frames to raw video before computing the
173 The output of the muxer consists of a line for each audio and video
176 @var{stream_index}, @var{packet_dts}, @var{packet_pts}, @var{packet_duration}, @var{packet_size}, 0x@var{CRC}
179 @var{CRC} is a hexadecimal number 0-padded to 8 digits containing the
184 For example to compute the CRC of the audio and video frames in
185 @file{INPUT}, converted to raw audio and video packets, and store it
186 in the file @file{out.crc}:
188 ffmpeg -i INPUT -f framecrc out.crc
191 To print the information to stdout, use the command:
193 ffmpeg -i INPUT -f framecrc -
196 With @command{ffmpeg}, you can select the output format to which the
197 audio and video frames are encoded before computing the CRC for each
198 packet by specifying the audio and video codec. For example, to
199 compute the CRC of each decoded input audio frame converted to PCM
200 unsigned 8-bit and of each decoded input video frame converted to
201 MPEG-2 video, use the command:
203 ffmpeg -i INPUT -c:a pcm_u8 -c:v mpeg2video -f framecrc -
206 See also the @ref{crc} muxer.
211 Per-packet hash testing format.
213 This muxer computes and prints a cryptographic hash for each audio
214 and video packet. This can be used for packet-by-packet equality
215 checks without having to individually do a binary comparison on each.
217 By default audio frames are converted to signed 16-bit raw audio and
218 video frames to raw video before computing the hash, but the output
219 of explicit conversions to other codecs can also be used. It uses the
220 SHA-256 cryptographic hash function by default, but supports several
223 The output of the muxer consists of a line for each audio and video
226 @var{stream_index}, @var{packet_dts}, @var{packet_pts}, @var{packet_duration}, @var{packet_size}, @var{hash}
229 @var{hash} is a hexadecimal number representing the computed hash
233 @item hash @var{algorithm}
234 Use the cryptographic hash function specified by the string @var{algorithm}.
235 Supported values include @code{MD5}, @code{murmur3}, @code{RIPEMD128},
236 @code{RIPEMD160}, @code{RIPEMD256}, @code{RIPEMD320}, @code{SHA160},
237 @code{SHA224}, @code{SHA256} (default), @code{SHA512/224}, @code{SHA512/256},
238 @code{SHA384}, @code{SHA512}, @code{CRC32} and @code{adler32}.
244 To compute the SHA-256 hash of the audio and video frames in @file{INPUT},
245 converted to raw audio and video packets, and store it in the file
248 ffmpeg -i INPUT -f framehash out.sha256
251 To print the information to stdout, using the MD5 hash function, use
254 ffmpeg -i INPUT -f framehash -hash md5 -
257 See also the @ref{hash} muxer.
262 Per-packet MD5 testing format.
264 This is a variant of the @ref{framehash} muxer. Unlike that muxer,
265 it defaults to using the MD5 hash function.
269 To compute the MD5 hash of the audio and video frames in @file{INPUT},
270 converted to raw audio and video packets, and store it in the file
273 ffmpeg -i INPUT -f framemd5 out.md5
276 To print the information to stdout, use the command:
278 ffmpeg -i INPUT -f framemd5 -
281 See also the @ref{framehash} and @ref{md5} muxers.
288 It accepts the following options:
292 Set the number of times to loop the output. Use @code{-1} for no loop, @code{0}
293 for looping indefinitely (default).
296 Force the delay (expressed in centiseconds) after the last frame. Each frame
297 ends with a delay until the next frame. The default is @code{-1}, which is a
298 special value to tell the muxer to re-use the previous delay. In case of a
299 loop, you might want to customize this value to mark a pause for instance.
302 For example, to encode a gif looping 10 times, with a 5 seconds delay between
305 ffmpeg -i INPUT -loop 10 -final_delay 500 out.gif
308 Note 1: if you wish to extract the frames into separate GIF files, you need to
309 force the @ref{image2} muxer:
311 ffmpeg -i INPUT -c:v gif -f image2 "out%d.gif"
314 Note 2: the GIF format has a very large time base: the delay between two frames
315 can therefore not be smaller than one centi second.
322 This muxer computes and prints a cryptographic hash of all the input
323 audio and video frames. This can be used for equality checks without
324 having to do a complete binary comparison.
326 By default audio frames are converted to signed 16-bit raw audio and
327 video frames to raw video before computing the hash, but the output
328 of explicit conversions to other codecs can also be used. Timestamps
329 are ignored. It uses the SHA-256 cryptographic hash function by default,
330 but supports several other algorithms.
332 The output of the muxer consists of a single line of the form:
333 @var{algo}=@var{hash}, where @var{algo} is a short string representing
334 the hash function used, and @var{hash} is a hexadecimal number
335 representing the computed hash.
338 @item hash @var{algorithm}
339 Use the cryptographic hash function specified by the string @var{algorithm}.
340 Supported values include @code{MD5}, @code{murmur3}, @code{RIPEMD128},
341 @code{RIPEMD160}, @code{RIPEMD256}, @code{RIPEMD320}, @code{SHA160},
342 @code{SHA224}, @code{SHA256} (default), @code{SHA512/224}, @code{SHA512/256},
343 @code{SHA384}, @code{SHA512}, @code{CRC32} and @code{adler32}.
349 To compute the SHA-256 hash of the input converted to raw audio and
350 video, and store it in the file @file{out.sha256}:
352 ffmpeg -i INPUT -f hash out.sha256
355 To print an MD5 hash to stdout use the command:
357 ffmpeg -i INPUT -f hash -hash md5 -
360 See also the @ref{framehash} muxer.
365 Apple HTTP Live Streaming muxer that segments MPEG-TS according to
366 the HTTP Live Streaming (HLS) specification.
368 It creates a playlist file, and one or more segment files. The output filename
369 specifies the playlist filename.
371 By default, the muxer creates a file for each segment produced. These files
372 have the same name as the playlist, followed by a sequential number and a
375 For example, to convert an input file with @command{ffmpeg}:
377 ffmpeg -i in.nut out.m3u8
379 This example will produce the playlist, @file{out.m3u8}, and segment files:
380 @file{out0.ts}, @file{out1.ts}, @file{out2.ts}, etc.
382 See also the @ref{segment} muxer, which provides a more generic and
383 flexible implementation of a segmenter, and can be used to perform HLS
388 This muxer supports the following options:
391 @item hls_init_time @var{seconds}
392 Set the initial target segment length in seconds. Default value is @var{0}.
393 Segment will be cut on the next key frame after this time has passed on the first m3u8 list.
394 After the initial playlist is filled @command{ffmpeg} will cut segments
395 at duration equal to @code{hls_time}
397 @item hls_time @var{seconds}
398 Set the target segment length in seconds. Default value is 2.
399 Segment will be cut on the next key frame after this time has passed.
401 @item hls_list_size @var{size}
402 Set the maximum number of playlist entries. If set to 0 the list file
403 will contain all the segments. Default value is 5.
405 @item hls_ts_options @var{options_list}
406 Set output format options using a :-separated list of key=value
407 parameters. Values containing @code{:} special characters must be
410 @item hls_wrap @var{wrap}
411 Set the number after which the segment filename number (the number
412 specified in each segment file) wraps. If set to 0 the number will be
413 never wrapped. Default value is 0.
415 This option is useful to avoid to fill the disk with many segment
416 files, and limits the maximum number of segment files written to disk
419 @item start_number @var{number}
420 Start the playlist sequence number from @var{number}. Default value is
423 @item hls_allow_cache @var{allowcache}
424 Explicitly set whether the client MAY (1) or MUST NOT (0) cache media segments.
426 @item hls_base_url @var{baseurl}
427 Append @var{baseurl} to every entry in the playlist.
428 Useful to generate playlists with absolute paths.
430 Note that the playlist sequence number must be unique for each segment
431 and it is not to be confused with the segment filename sequence number
432 which can be cyclic, for example if the @option{wrap} option is
435 @item hls_segment_filename @var{filename}
436 Set the segment filename. Unless @code{hls_flags single_file} is set,
437 @var{filename} is used as a string format with the segment number:
439 ffmpeg -i in.nut -hls_segment_filename 'file%03d.ts' out.m3u8
441 This example will produce the playlist, @file{out.m3u8}, and segment files:
442 @file{file000.ts}, @file{file001.ts}, @file{file002.ts}, etc.
445 Use strftime on @var{filename} to expand the segment filename with localtime.
446 The segment number is also available in this mode, but to use it, you need to specify second_level_segment_index
447 hls_flag and %%d will be the specifier.
449 ffmpeg -i in.nut -use_localtime 1 -hls_segment_filename 'file-%Y%m%d-%s.ts' out.m3u8
451 This example will produce the playlist, @file{out.m3u8}, and segment files:
452 @file{file-20160215-1455569023.ts}, @file{file-20160215-1455569024.ts}, etc.
454 ffmpeg -i in.nut -use_localtime 1 -hls_flags second_level_segment_index -hls_segment_filename 'file-%Y%m%d-%%04d.ts' out.m3u8
456 This example will produce the playlist, @file{out.m3u8}, and segment files:
457 @file{file-20160215-0001.ts}, @file{file-20160215-0002.ts}, etc.
459 @item use_localtime_mkdir
460 Used together with -use_localtime, it will create up to one subdirectory which
461 is expanded in @var{filename}.
463 ffmpeg -i in.nut -use_localtime 1 -use_localtime_mkdir 1 -hls_segment_filename '%Y%m%d/file-%Y%m%d-%s.ts' out.m3u8
465 This example will create a directory 201560215 (if it does not exist), and then
466 produce the playlist, @file{out.m3u8}, and segment files:
467 @file{201560215/file-20160215-1455569023.ts}, @file{201560215/file-20160215-1455569024.ts}, etc.
470 @item hls_key_info_file @var{key_info_file}
471 Use the information in @var{key_info_file} for segment encryption. The first
472 line of @var{key_info_file} specifies the key URI written to the playlist. The
473 key URL is used to access the encryption key during playback. The second line
474 specifies the path to the key file used to obtain the key during the encryption
475 process. The key file is read as a single packed array of 16 octets in binary
476 format. The optional third line specifies the initialization vector (IV) as a
477 hexadecimal string to be used instead of the segment sequence number (default)
478 for encryption. Changes to @var{key_info_file} will result in segment
479 encryption with the new key/IV and an entry in the playlist for the new key
482 Key info file format:
491 http://server/file.key
496 Example key file paths:
504 0123456789ABCDEF0123456789ABCDEF
507 Key info file example:
509 http://server/file.key
511 0123456789ABCDEF0123456789ABCDEF
514 Example shell script:
518 openssl rand 16 > file.key
519 echo $BASE_URL/file.key > file.keyinfo
520 echo file.key >> file.keyinfo
521 echo $(openssl rand -hex 16) >> file.keyinfo
522 ffmpeg -f lavfi -re -i testsrc -c:v h264 -hls_flags delete_segments \
523 -hls_key_info_file file.keyinfo out.m3u8
526 @item hls_flags single_file
527 If this flag is set, the muxer will store all segments in a single MPEG-TS
528 file, and will use byte ranges in the playlist. HLS playlists generated with
529 this way will have the version number 4.
532 ffmpeg -i in.nut -hls_flags single_file out.m3u8
534 Will produce the playlist, @file{out.m3u8}, and a single segment file,
537 @item hls_flags delete_segments
538 Segment files removed from the playlist are deleted after a period of time
539 equal to the duration of the segment plus the duration of the playlist.
541 @item hls_flags append_list
542 Append new segments into the end of old segment list,
543 and remove the @code{#EXT-X-ENDLIST} from the old segment list.
545 @item hls_flags round_durations
546 Round the duration info in the playlist file segment info to integer
547 values, instead of using floating point.
549 @item hls_flags discont_starts
550 Add the @code{#EXT-X-DISCONTINUITY} tag to the playlist, before the
551 first segment's information.
553 @item hls_flags omit_endlist
554 Do not append the @code{EXT-X-ENDLIST} tag at the end of the playlist.
556 @item hls_flags split_by_time
557 Allow segments to start on frames other than keyframes. This improves
558 behavior on some players when the time between keyframes is inconsistent,
559 but may make things worse on others, and can cause some oddities during
560 seeking. This flag should be used with the @code{hls_time} option.
562 @item hls_flags program_date_time
563 Generate @code{EXT-X-PROGRAM-DATE-TIME} tags.
565 @item hls_flags second_level_segment_index
566 Makes it possible to use segment indexes as %%d besides date/time values when use_localtime is on.
568 @item hls_playlist_type event
569 Emit @code{#EXT-X-PLAYLIST-TYPE:EVENT} in the m3u8 header. Forces
570 @option{hls_list_size} to 0; the playlist can only be appended to.
572 @item hls_playlist_type vod
573 Emit @code{#EXT-X-PLAYLIST-TYPE:VOD} in the m3u8 header. Forces
574 @option{hls_list_size} to 0; the playlist must not change.
577 Use the given HTTP method to create the hls files.
579 ffmpeg -re -i in.ts -f hls -method PUT http://example.com/live/out.m3u8
581 This example will upload all the mpegts segment files to the HTTP
582 server using the HTTP PUT method, and update the m3u8 files every
583 @code{refresh} times using the same method.
584 Note that the HTTP server must support the given method for uploading
593 Microsoft's icon file format (ICO) has some strict limitations that should be noted:
597 Size cannot exceed 256 pixels in any dimension
600 Only BMP and PNG images can be stored
603 If a BMP image is used, it must be one of the following pixel formats:
605 BMP Bit Depth FFmpeg Pixel Format
615 If a BMP image is used, it must use the BITMAPINFOHEADER DIB header
618 If a PNG image is used, it must use the rgba pixel format
626 The image file muxer writes video frames to image files.
628 The output filenames are specified by a pattern, which can be used to
629 produce sequentially numbered series of files.
630 The pattern may contain the string "%d" or "%0@var{N}d", this string
631 specifies the position of the characters representing a numbering in
632 the filenames. If the form "%0@var{N}d" is used, the string
633 representing the number in each filename is 0-padded to @var{N}
634 digits. The literal character '%' can be specified in the pattern with
637 If the pattern contains "%d" or "%0@var{N}d", the first filename of
638 the file list specified will contain the number 1, all the following
639 numbers will be sequential.
641 The pattern may contain a suffix which is used to automatically
642 determine the format of the image files to write.
644 For example the pattern "img-%03d.bmp" will specify a sequence of
645 filenames of the form @file{img-001.bmp}, @file{img-002.bmp}, ...,
646 @file{img-010.bmp}, etc.
647 The pattern "img%%-%d.jpg" will specify a sequence of filenames of the
648 form @file{img%-1.jpg}, @file{img%-2.jpg}, ..., @file{img%-10.jpg},
653 The following example shows how to use @command{ffmpeg} for creating a
654 sequence of files @file{img-001.jpeg}, @file{img-002.jpeg}, ...,
655 taking one image every second from the input video:
657 ffmpeg -i in.avi -vsync cfr -r 1 -f image2 'img-%03d.jpeg'
660 Note that with @command{ffmpeg}, if the format is not specified with the
661 @code{-f} option and the output filename specifies an image file
662 format, the image2 muxer is automatically selected, so the previous
663 command can be written as:
665 ffmpeg -i in.avi -vsync cfr -r 1 'img-%03d.jpeg'
668 Note also that the pattern must not necessarily contain "%d" or
669 "%0@var{N}d", for example to create a single image file
670 @file{img.jpeg} from the start of the input video you can employ the command:
672 ffmpeg -i in.avi -f image2 -frames:v 1 img.jpeg
675 The @option{strftime} option allows you to expand the filename with
676 date and time information. Check the documentation of
677 the @code{strftime()} function for the syntax.
679 For example to generate image files from the @code{strftime()}
680 "%Y-%m-%d_%H-%M-%S" pattern, the following @command{ffmpeg} command
683 ffmpeg -f v4l2 -r 1 -i /dev/video0 -f image2 -strftime 1 "%Y-%m-%d_%H-%M-%S.jpg"
690 Start the sequence from the specified number. Default value is 0.
693 If set to 1, the filename will always be interpreted as just a
694 filename, not a pattern, and the corresponding file will be continuously
695 overwritten with new images. Default value is 0.
698 If set to 1, expand the filename with date and time information from
699 @code{strftime()}. Default value is 0.
702 The image muxer supports the .Y.U.V image file format. This format is
703 special in that that each image frame consists of three files, for
704 each of the YUV420P components. To read or write this image file format,
705 specify the name of the '.Y' file. The muxer will automatically open the
706 '.U' and '.V' files as required.
710 Matroska container muxer.
712 This muxer implements the matroska and webm container specs.
716 The recognized metadata settings in this muxer are:
720 Set title name provided to a single track.
723 Specify the language of the track in the Matroska languages form.
725 The language can be either the 3 letters bibliographic ISO-639-2 (ISO
726 639-2/B) form (like "fre" for French), or a language code mixed with a
727 country code for specialities in languages (like "fre-ca" for Canadian
731 Set stereo 3D video layout of two views in a single video track.
733 The following values are recognized:
738 Both views are arranged side by side, Left-eye view is on the left
740 Both views are arranged in top-bottom orientation, Left-eye view is at bottom
742 Both views are arranged in top-bottom orientation, Left-eye view is on top
743 @item checkerboard_rl
744 Each view is arranged in a checkerboard interleaved pattern, Left-eye view being first
745 @item checkerboard_lr
746 Each view is arranged in a checkerboard interleaved pattern, Right-eye view being first
747 @item row_interleaved_rl
748 Each view is constituted by a row based interleaving, Right-eye view is first row
749 @item row_interleaved_lr
750 Each view is constituted by a row based interleaving, Left-eye view is first row
751 @item col_interleaved_rl
752 Both views are arranged in a column based interleaving manner, Right-eye view is first column
753 @item col_interleaved_lr
754 Both views are arranged in a column based interleaving manner, Left-eye view is first column
755 @item anaglyph_cyan_red
756 All frames are in anaglyph format viewable through red-cyan filters
758 Both views are arranged side by side, Right-eye view is on the left
759 @item anaglyph_green_magenta
760 All frames are in anaglyph format viewable through green-magenta filters
762 Both eyes laced in one Block, Left-eye view is first
764 Both eyes laced in one Block, Right-eye view is first
768 For example a 3D WebM clip can be created using the following command line:
770 ffmpeg -i sample_left_right_clip.mpg -an -c:v libvpx -metadata stereo_mode=left_right -y stereo_clip.webm
775 This muxer supports the following options:
778 @item reserve_index_space
779 By default, this muxer writes the index for seeking (called cues in Matroska
780 terms) at the end of the file, because it cannot know in advance how much space
781 to leave for the index at the beginning of the file. However for some use cases
782 -- e.g. streaming where seeking is possible but slow -- it is useful to put the
783 index at the beginning of the file.
785 If this option is set to a non-zero value, the muxer will reserve a given amount
786 of space in the file header and then try to write the cues there when the muxing
787 finishes. If the available space does not suffice, muxing will fail. A safe size
788 for most use cases should be about 50kB per hour of video.
790 Note that cues are only written if the output is seekable and this option will
791 have no effect if it is not.
799 This is a variant of the @ref{hash} muxer. Unlike that muxer, it
800 defaults to using the MD5 hash function.
804 To compute the MD5 hash of the input converted to raw
805 audio and video, and store it in the file @file{out.md5}:
807 ffmpeg -i INPUT -f md5 out.md5
810 You can print the MD5 to stdout with the command:
812 ffmpeg -i INPUT -f md5 -
815 See also the @ref{hash} and @ref{framemd5} muxers.
817 @section mov, mp4, ismv
819 MOV/MP4/ISMV (Smooth Streaming) muxer.
821 The mov/mp4/ismv muxer supports fragmentation. Normally, a MOV/MP4
822 file has all the metadata about all packets stored in one location
823 (written at the end of the file, it can be moved to the start for
824 better playback by adding @var{faststart} to the @var{movflags}, or
825 using the @command{qt-faststart} tool). A fragmented
826 file consists of a number of fragments, where packets and metadata
827 about these packets are stored together. Writing a fragmented
828 file has the advantage that the file is decodable even if the
829 writing is interrupted (while a normal MOV/MP4 is undecodable if
830 it is not properly finished), and it requires less memory when writing
831 very long files (since writing normal MOV/MP4 files stores info about
832 every single packet in memory until the file is closed). The downside
833 is that it is less compatible with other applications.
837 Fragmentation is enabled by setting one of the AVOptions that define
838 how to cut the file into fragments:
841 @item -moov_size @var{bytes}
842 Reserves space for the moov atom at the beginning of the file instead of placing the
843 moov atom at the end. If the space reserved is insufficient, muxing will fail.
844 @item -movflags frag_keyframe
845 Start a new fragment at each video keyframe.
846 @item -frag_duration @var{duration}
847 Create fragments that are @var{duration} microseconds long.
848 @item -frag_size @var{size}
849 Create fragments that contain up to @var{size} bytes of payload data.
850 @item -movflags frag_custom
851 Allow the caller to manually choose when to cut fragments, by
852 calling @code{av_write_frame(ctx, NULL)} to write a fragment with
853 the packets written so far. (This is only useful with other
854 applications integrating libavformat, not from @command{ffmpeg}.)
855 @item -min_frag_duration @var{duration}
856 Don't create fragments that are shorter than @var{duration} microseconds long.
859 If more than one condition is specified, fragments are cut when
860 one of the specified conditions is fulfilled. The exception to this is
861 @code{-min_frag_duration}, which has to be fulfilled for any of the other
864 Additionally, the way the output file is written can be adjusted
865 through a few other options:
868 @item -movflags empty_moov
869 Write an initial moov atom directly at the start of the file, without
870 describing any samples in it. Generally, an mdat/moov pair is written
871 at the start of the file, as a normal MOV/MP4 file, containing only
872 a short portion of the file. With this option set, there is no initial
873 mdat atom, and the moov atom only describes the tracks but has
876 This option is implicitly set when writing ismv (Smooth Streaming) files.
877 @item -movflags separate_moof
878 Write a separate moof (movie fragment) atom for each track. Normally,
879 packets for all tracks are written in a moof atom (which is slightly
880 more efficient), but with this option set, the muxer writes one moof/mdat
881 pair for each track, making it easier to separate tracks.
883 This option is implicitly set when writing ismv (Smooth Streaming) files.
884 @item -movflags faststart
885 Run a second pass moving the index (moov atom) to the beginning of the file.
886 This operation can take a while, and will not work in various situations such
887 as fragmented output, thus it is not enabled by default.
888 @item -movflags rtphint
889 Add RTP hinting tracks to the output file.
890 @item -movflags disable_chpl
891 Disable Nero chapter markers (chpl atom). Normally, both Nero chapters
892 and a QuickTime chapter track are written to the file. With this option
893 set, only the QuickTime chapter track will be written. Nero chapters can
894 cause failures when the file is reprocessed with certain tagging programs, like
895 mp3Tag 2.61a and iTunes 11.3, most likely other versions are affected as well.
896 @item -movflags omit_tfhd_offset
897 Do not write any absolute base_data_offset in tfhd atoms. This avoids
898 tying fragments to absolute byte positions in the file/streams.
899 @item -movflags default_base_moof
900 Similarly to the omit_tfhd_offset, this flag avoids writing the
901 absolute base_data_offset field in tfhd atoms, but does so by using
902 the new default-base-is-moof flag instead. This flag is new from
903 14496-12:2012. This may make the fragments easier to parse in certain
904 circumstances (avoiding basing track fragment location calculations
905 on the implicit end of the previous track fragment).
907 Specify @code{on} to force writing a timecode track, @code{off} to disable it
908 and @code{auto} to write a timecode track only for mov and mp4 output (default).
913 Smooth Streaming content can be pushed in real time to a publishing
914 point on IIS with this muxer. Example:
916 ffmpeg -re @var{<normal input/transcoding options>} -movflags isml+frag_keyframe -f ismv http://server/publishingpoint.isml/Streams(Encoder1)
919 @subsection Audible AAX
921 Audible AAX files are encrypted M4B files, and they can be decrypted by specifying a 4 byte activation secret.
923 ffmpeg -activation_bytes 1CEB00DA -i test.aax -vn -c:a copy output.mp4
928 The MP3 muxer writes a raw MP3 stream with the following optional features:
931 An ID3v2 metadata header at the beginning (enabled by default). Versions 2.3 and
932 2.4 are supported, the @code{id3v2_version} private option controls which one is
933 used (3 or 4). Setting @code{id3v2_version} to 0 disables the ID3v2 header
936 The muxer supports writing attached pictures (APIC frames) to the ID3v2 header.
937 The pictures are supplied to the muxer in form of a video stream with a single
938 packet. There can be any number of those streams, each will correspond to a
939 single APIC frame. The stream metadata tags @var{title} and @var{comment} map
940 to APIC @var{description} and @var{picture type} respectively. See
941 @url{http://id3.org/id3v2.4.0-frames} for allowed picture types.
943 Note that the APIC frames must be written at the beginning, so the muxer will
944 buffer the audio frames until it gets all the pictures. It is therefore advised
945 to provide the pictures as soon as possible to avoid excessive buffering.
948 A Xing/LAME frame right after the ID3v2 header (if present). It is enabled by
949 default, but will be written only if the output is seekable. The
950 @code{write_xing} private option can be used to disable it. The frame contains
951 various information that may be useful to the decoder, like the audio duration
955 A legacy ID3v1 tag at the end of the file (disabled by default). It may be
956 enabled with the @code{write_id3v1} private option, but as its capabilities are
957 very limited, its usage is not recommended.
962 Write an mp3 with an ID3v2.3 header and an ID3v1 footer:
964 ffmpeg -i INPUT -id3v2_version 3 -write_id3v1 1 out.mp3
967 To attach a picture to an mp3 file select both the audio and the picture stream
970 ffmpeg -i input.mp3 -i cover.png -c copy -map 0 -map 1
971 -metadata:s:v title="Album cover" -metadata:s:v comment="Cover (Front)" out.mp3
974 Write a "clean" MP3 without any extra features:
976 ffmpeg -i input.wav -write_xing 0 -id3v2_version 0 out.mp3
981 MPEG transport stream muxer.
983 This muxer implements ISO 13818-1 and part of ETSI EN 300 468.
985 The recognized metadata settings in mpegts muxer are @code{service_provider}
986 and @code{service_name}. If they are not set the default for
987 @code{service_provider} is "FFmpeg" and the default for
988 @code{service_name} is "Service01".
992 The muxer options are:
995 @item mpegts_original_network_id @var{number}
996 Set the original_network_id (default 0x0001). This is unique identifier
997 of a network in DVB. Its main use is in the unique identification of a
998 service through the path Original_Network_ID, Transport_Stream_ID.
999 @item mpegts_transport_stream_id @var{number}
1000 Set the transport_stream_id (default 0x0001). This identifies a
1002 @item mpegts_service_id @var{number}
1003 Set the service_id (default 0x0001) also known as program in DVB.
1004 @item mpegts_service_type @var{number}
1005 Set the program service_type (default @var{digital_tv}), see below
1006 a list of pre defined values.
1007 @item mpegts_pmt_start_pid @var{number}
1008 Set the first PID for PMT (default 0x1000, max 0x1f00).
1009 @item mpegts_start_pid @var{number}
1010 Set the first PID for data packets (default 0x0100, max 0x0f00).
1011 @item mpegts_m2ts_mode @var{number}
1012 Enable m2ts mode if set to 1. Default value is -1 which disables m2ts mode.
1013 @item muxrate @var{number}
1014 Set a constant muxrate (default VBR).
1015 @item pcr_period @var{numer}
1016 Override the default PCR retransmission time (default 20ms), ignored
1017 if variable muxrate is selected.
1018 @item pat_period @var{number}
1019 Maximal time in seconds between PAT/PMT tables.
1020 @item sdt_period @var{number}
1021 Maximal time in seconds between SDT tables.
1022 @item pes_payload_size @var{number}
1023 Set minimum PES packet payload in bytes.
1024 @item mpegts_flags @var{flags}
1025 Set flags (see below).
1026 @item mpegts_copyts @var{number}
1027 Preserve original timestamps, if value is set to 1. Default value is -1, which
1028 results in shifting timestamps so that they start from 0.
1029 @item tables_version @var{number}
1030 Set PAT, PMT and SDT version (default 0, valid values are from 0 to 31, inclusively).
1031 This option allows updating stream structure so that standard consumer may
1032 detect the change. To do so, reopen output AVFormatContext (in case of API
1033 usage) or restart ffmpeg instance, cyclically changing tables_version value:
1035 ffmpeg -i source1.ts -codec copy -f mpegts -tables_version 0 udp://1.1.1.1:1111
1036 ffmpeg -i source2.ts -codec copy -f mpegts -tables_version 1 udp://1.1.1.1:1111
1038 ffmpeg -i source3.ts -codec copy -f mpegts -tables_version 31 udp://1.1.1.1:1111
1039 ffmpeg -i source1.ts -codec copy -f mpegts -tables_version 0 udp://1.1.1.1:1111
1040 ffmpeg -i source2.ts -codec copy -f mpegts -tables_version 1 udp://1.1.1.1:1111
1045 Option @option{mpegts_service_type} accepts the following values:
1049 Any hexdecimal value between 0x01 to 0xff as defined in ETSI 300 468.
1053 Digital Radio service.
1056 @item advanced_codec_digital_radio
1057 Advanced Codec Digital Radio service.
1058 @item mpeg2_digital_hdtv
1059 MPEG2 Digital HDTV service.
1060 @item advanced_codec_digital_sdtv
1061 Advanced Codec Digital SDTV service.
1062 @item advanced_codec_digital_hdtv
1063 Advanced Codec Digital HDTV service.
1066 Option @option{mpegts_flags} may take a set of such flags:
1069 @item resend_headers
1070 Reemit PAT/PMT before writing the next packet.
1072 Use LATM packetization for AAC.
1073 @item pat_pmt_at_frames
1074 Reemit PAT and PMT at each video frame.
1076 Conform to System B (DVB) instead of System A (ATSC).
1077 @item initial_discontinuity
1078 Mark the initial packet of each stream as discontinuity.
1084 ffmpeg -i file.mpg -c copy \
1085 -mpegts_original_network_id 0x1122 \
1086 -mpegts_transport_stream_id 0x3344 \
1087 -mpegts_service_id 0x5566 \
1088 -mpegts_pmt_start_pid 0x1500 \
1089 -mpegts_start_pid 0x150 \
1090 -metadata service_provider="Some provider" \
1091 -metadata service_name="Some Channel" \
1095 @section mxf, mxf_d10
1101 The muxer options are:
1104 @item store_user_comments @var{bool}
1105 Set if user comments should be stored if available or never.
1106 IRT D-10 does not allow user comments. The default is thus to write them for
1107 mxf but not for mxf_d10
1114 This muxer does not generate any output file, it is mainly useful for
1115 testing or benchmarking purposes.
1117 For example to benchmark decoding with @command{ffmpeg} you can use the
1120 ffmpeg -benchmark -i INPUT -f null out.null
1123 Note that the above command does not read or write the @file{out.null}
1124 file, but specifying the output file is required by the @command{ffmpeg}
1127 Alternatively you can write the command as:
1129 ffmpeg -benchmark -i INPUT -f null -
1135 @item -syncpoints @var{flags}
1136 Change the syncpoint usage in nut:
1138 @item @var{default} use the normal low-overhead seeking aids.
1139 @item @var{none} do not use the syncpoints at all, reducing the overhead but making the stream non-seekable;
1140 Use of this option is not recommended, as the resulting files are very damage
1141 sensitive and seeking is not possible. Also in general the overhead from
1142 syncpoints is negligible. Note, -@code{write_index} 0 can be used to disable
1143 all growing data tables, allowing to mux endless streams with limited memory
1144 and without these disadvantages.
1145 @item @var{timestamped} extend the syncpoint with a wallclock field.
1147 The @var{none} and @var{timestamped} flags are experimental.
1148 @item -write_index @var{bool}
1149 Write index at the end, the default is to write an index.
1153 ffmpeg -i INPUT -f_strict experimental -syncpoints none - | processor
1158 Ogg container muxer.
1161 @item -page_duration @var{duration}
1162 Preferred page duration, in microseconds. The muxer will attempt to create
1163 pages that are approximately @var{duration} microseconds long. This allows the
1164 user to compromise between seek granularity and container overhead. The default
1165 is 1 second. A value of 0 will fill all segments, making pages as large as
1166 possible. A value of 1 will effectively use 1 packet-per-page in most
1167 situations, giving a small seek granularity at the cost of additional container
1169 @item -serial_offset @var{value}
1170 Serial value from which to set the streams serial number.
1171 Setting it to different and sufficiently large values ensures that the produced
1172 ogg files can be safely chained.
1177 @section segment, stream_segment, ssegment
1179 Basic stream segmenter.
1181 This muxer outputs streams to a number of separate files of nearly
1182 fixed duration. Output filename pattern can be set in a fashion
1183 similar to @ref{image2}, or by using a @code{strftime} template if
1184 the @option{strftime} option is enabled.
1186 @code{stream_segment} is a variant of the muxer used to write to
1187 streaming output formats, i.e. which do not require global headers,
1188 and is recommended for outputting e.g. to MPEG transport stream segments.
1189 @code{ssegment} is a shorter alias for @code{stream_segment}.
1191 Every segment starts with a keyframe of the selected reference stream,
1192 which is set through the @option{reference_stream} option.
1194 Note that if you want accurate splitting for a video file, you need to
1195 make the input key frames correspond to the exact splitting times
1196 expected by the segmenter, or the segment muxer will start the new
1197 segment with the key frame found next after the specified start
1200 The segment muxer works best with a single constant frame rate video.
1202 Optionally it can generate a list of the created segments, by setting
1203 the option @var{segment_list}. The list type is specified by the
1204 @var{segment_list_type} option. The entry filenames in the segment
1205 list are set by default to the basename of the corresponding segment
1208 See also the @ref{hls} muxer, which provides a more specific
1209 implementation for HLS segmentation.
1213 The segment muxer supports the following options:
1216 @item increment_tc @var{1|0}
1217 if set to @code{1}, increment timecode between each segment
1218 If this is selected, the input need to have
1219 a timecode in the first video stream. Default value is
1222 @item reference_stream @var{specifier}
1223 Set the reference stream, as specified by the string @var{specifier}.
1224 If @var{specifier} is set to @code{auto}, the reference is chosen
1225 automatically. Otherwise it must be a stream specifier (see the ``Stream
1226 specifiers'' chapter in the ffmpeg manual) which specifies the
1227 reference stream. The default value is @code{auto}.
1229 @item segment_format @var{format}
1230 Override the inner container format, by default it is guessed by the filename
1233 @item segment_format_options @var{options_list}
1234 Set output format options using a :-separated list of key=value
1235 parameters. Values containing the @code{:} special character must be
1238 @item segment_list @var{name}
1239 Generate also a listfile named @var{name}. If not specified no
1240 listfile is generated.
1242 @item segment_list_flags @var{flags}
1243 Set flags affecting the segment list generation.
1245 It currently supports the following flags:
1248 Allow caching (only affects M3U8 list files).
1251 Allow live-friendly file generation.
1254 @item segment_list_size @var{size}
1255 Update the list file so that it contains at most @var{size}
1256 segments. If 0 the list file will contain all the segments. Default
1259 @item segment_list_entry_prefix @var{prefix}
1260 Prepend @var{prefix} to each entry. Useful to generate absolute paths.
1261 By default no prefix is applied.
1263 @item segment_list_type @var{type}
1264 Select the listing format.
1266 The following values are recognized:
1269 Generate a flat list for the created segments, one segment per line.
1272 Generate a list for the created segments, one segment per line,
1273 each line matching the format (comma-separated values):
1275 @var{segment_filename},@var{segment_start_time},@var{segment_end_time}
1278 @var{segment_filename} is the name of the output file generated by the
1279 muxer according to the provided pattern. CSV escaping (according to
1280 RFC4180) is applied if required.
1282 @var{segment_start_time} and @var{segment_end_time} specify
1283 the segment start and end time expressed in seconds.
1285 A list file with the suffix @code{".csv"} or @code{".ext"} will
1286 auto-select this format.
1288 @samp{ext} is deprecated in favor or @samp{csv}.
1291 Generate an ffconcat file for the created segments. The resulting file
1292 can be read using the FFmpeg @ref{concat} demuxer.
1294 A list file with the suffix @code{".ffcat"} or @code{".ffconcat"} will
1295 auto-select this format.
1298 Generate an extended M3U8 file, version 3, compliant with
1299 @url{http://tools.ietf.org/id/draft-pantos-http-live-streaming}.
1301 A list file with the suffix @code{".m3u8"} will auto-select this format.
1304 If not specified the type is guessed from the list file name suffix.
1306 @item segment_time @var{time}
1307 Set segment duration to @var{time}, the value must be a duration
1308 specification. Default value is "2". See also the
1309 @option{segment_times} option.
1311 Note that splitting may not be accurate, unless you force the
1312 reference stream key-frames at the given time. See the introductory
1313 notice and the examples below.
1315 @item segment_atclocktime @var{1|0}
1316 If set to "1" split at regular clock time intervals starting from 00:00
1317 o'clock. The @var{time} value specified in @option{segment_time} is
1318 used for setting the length of the splitting interval.
1320 For example with @option{segment_time} set to "900" this makes it possible
1321 to create files at 12:00 o'clock, 12:15, 12:30, etc.
1323 Default value is "0".
1325 @item segment_clocktime_offset @var{duration}
1326 Delay the segment splitting times with the specified duration when using
1327 @option{segment_atclocktime}.
1329 For example with @option{segment_time} set to "900" and
1330 @option{segment_clocktime_offset} set to "300" this makes it possible to
1331 create files at 12:05, 12:20, 12:35, etc.
1333 Default value is "0".
1335 @item segment_clocktime_wrap_duration @var{duration}
1336 Force the segmenter to only start a new segment if a packet reaches the muxer
1337 within the specified duration after the segmenting clock time. This way you
1338 can make the segmenter more resilient to backward local time jumps, such as
1339 leap seconds or transition to standard time from daylight savings time.
1341 Assuming that the delay between the packets of your source is less than 0.5
1342 second you can detect a leap second by specifying 0.5 as the duration.
1344 Default is the maximum possible duration which means starting a new segment
1345 regardless of the elapsed time since the last clock time.
1347 @item segment_time_delta @var{delta}
1348 Specify the accuracy time when selecting the start time for a
1349 segment, expressed as a duration specification. Default value is "0".
1351 When delta is specified a key-frame will start a new segment if its
1352 PTS satisfies the relation:
1354 PTS >= start_time - time_delta
1357 This option is useful when splitting video content, which is always
1358 split at GOP boundaries, in case a key frame is found just before the
1359 specified split time.
1361 In particular may be used in combination with the @file{ffmpeg} option
1362 @var{force_key_frames}. The key frame times specified by
1363 @var{force_key_frames} may not be set accurately because of rounding
1364 issues, with the consequence that a key frame time may result set just
1365 before the specified time. For constant frame rate videos a value of
1366 1/(2*@var{frame_rate}) should address the worst case mismatch between
1367 the specified time and the time set by @var{force_key_frames}.
1369 @item segment_times @var{times}
1370 Specify a list of split points. @var{times} contains a list of comma
1371 separated duration specifications, in increasing order. See also
1372 the @option{segment_time} option.
1374 @item segment_frames @var{frames}
1375 Specify a list of split video frame numbers. @var{frames} contains a
1376 list of comma separated integer numbers, in increasing order.
1378 This option specifies to start a new segment whenever a reference
1379 stream key frame is found and the sequential number (starting from 0)
1380 of the frame is greater or equal to the next value in the list.
1382 @item segment_wrap @var{limit}
1383 Wrap around segment index once it reaches @var{limit}.
1385 @item segment_start_number @var{number}
1386 Set the sequence number of the first segment. Defaults to @code{0}.
1388 @item strftime @var{1|0}
1389 Use the @code{strftime} function to define the name of the new
1390 segments to write. If this is selected, the output segment name must
1391 contain a @code{strftime} function template. Default value is
1394 @item break_non_keyframes @var{1|0}
1395 If enabled, allow segments to start on frames other than keyframes. This
1396 improves behavior on some players when the time between keyframes is
1397 inconsistent, but may make things worse on others, and can cause some oddities
1398 during seeking. Defaults to @code{0}.
1400 @item reset_timestamps @var{1|0}
1401 Reset timestamps at the begin of each segment, so that each segment
1402 will start with near-zero timestamps. It is meant to ease the playback
1403 of the generated segments. May not work with some combinations of
1404 muxers/codecs. It is set to @code{0} by default.
1406 @item initial_offset @var{offset}
1407 Specify timestamp offset to apply to the output packet timestamps. The
1408 argument must be a time duration specification, and defaults to 0.
1410 @item write_empty_segments @var{1|0}
1411 If enabled, write an empty segment if there are no packets during the period a
1412 segment would usually span. Otherwise, the segment will be filled with the next
1413 packet written. Defaults to @code{0}.
1416 @subsection Examples
1420 Remux the content of file @file{in.mkv} to a list of segments
1421 @file{out-000.nut}, @file{out-001.nut}, etc., and write the list of
1422 generated segments to @file{out.list}:
1424 ffmpeg -i in.mkv -codec copy -map 0 -f segment -segment_list out.list out%03d.nut
1428 Segment input and set output format options for the output segments:
1430 ffmpeg -i in.mkv -f segment -segment_time 10 -segment_format_options movflags=+faststart out%03d.mp4
1434 Segment the input file according to the split points specified by the
1435 @var{segment_times} option:
1437 ffmpeg -i in.mkv -codec copy -map 0 -f segment -segment_list out.csv -segment_times 1,2,3,5,8,13,21 out%03d.nut
1441 Use the @command{ffmpeg} @option{force_key_frames}
1442 option to force key frames in the input at the specified location, together
1443 with the segment option @option{segment_time_delta} to account for
1444 possible roundings operated when setting key frame times.
1446 ffmpeg -i in.mkv -force_key_frames 1,2,3,5,8,13,21 -codec:v mpeg4 -codec:a pcm_s16le -map 0 \
1447 -f segment -segment_list out.csv -segment_times 1,2,3,5,8,13,21 -segment_time_delta 0.05 out%03d.nut
1449 In order to force key frames on the input file, transcoding is
1453 Segment the input file by splitting the input file according to the
1454 frame numbers sequence specified with the @option{segment_frames} option:
1456 ffmpeg -i in.mkv -codec copy -map 0 -f segment -segment_list out.csv -segment_frames 100,200,300,500,800 out%03d.nut
1460 Convert the @file{in.mkv} to TS segments using the @code{libx264}
1461 and @code{aac} encoders:
1463 ffmpeg -i in.mkv -map 0 -codec:v libx264 -codec:a aac -f ssegment -segment_list out.list out%03d.ts
1467 Segment the input file, and create an M3U8 live playlist (can be used
1468 as live HLS source):
1470 ffmpeg -re -i in.mkv -codec copy -map 0 -f segment -segment_list playlist.m3u8 \
1471 -segment_list_flags +live -segment_time 10 out%03d.mkv
1475 @section smoothstreaming
1477 Smooth Streaming muxer generates a set of files (Manifest, chunks) suitable for serving with conventional web server.
1481 Specify the number of fragments kept in the manifest. Default 0 (keep all).
1483 @item extra_window_size
1484 Specify the number of fragments kept outside of the manifest before removing from disk. Default 5.
1486 @item lookahead_count
1487 Specify the number of lookahead fragments. Default 2.
1489 @item min_frag_duration
1490 Specify the minimum fragment duration (in microseconds). Default 5000000.
1492 @item remove_at_exit
1493 Specify whether to remove all fragments when finished. Default 0 (do not remove).
1499 The fifo pseudo-muxer allows the separation of encoding and muxing by using
1500 first-in-first-out queue and running the actual muxer in a separate thread. This
1501 is especially useful in combination with the @ref{tee} muxer and can be used to
1502 send data to several destinations with different reliability/writing speed/latency.
1504 API users should be aware that callback functions (interrupt_callback,
1505 io_open and io_close) used within its AVFormatContext must be thread-safe.
1507 The behavior of the fifo muxer if the queue fills up or if the output fails is
1513 output can be transparently restarted with configurable delay between retries
1514 based on real time or time of the processed stream.
1517 encoding can be blocked during temporary failure, or continue transparently
1518 dropping packets in case fifo queue fills up.
1525 Specify the format name. Useful if it cannot be guessed from the
1529 Specify size of the queue (number of packets). Default value is 60.
1532 Specify format options for the underlying muxer. Muxer options can be specified
1533 as a list of @var{key}=@var{value} pairs separated by ':'.
1535 @item drop_pkts_on_overflow @var{bool}
1536 If set to 1 (true), in case the fifo queue fills up, packets will be dropped
1537 rather than blocking the encoder. This makes it possible to continue streaming without
1538 delaying the input, at the cost of omitting part of the stream. By default
1539 this option is set to 0 (false), so in such cases the encoder will be blocked
1540 until the muxer processes some of the packets and none of them is lost.
1542 @item attempt_recovery @var{bool}
1543 If failure occurs, attempt to recover the output. This is especially useful
1544 when used with network output, since it makes it possible to restart streaming transparently.
1545 By default this option is set to 0 (false).
1547 @item max_recovery_attempts
1548 Sets maximum number of successive unsuccessful recovery attempts after which
1549 the output fails permanently. By default this option is set to 0 (unlimited).
1551 @item recovery_wait_time @var{duration}
1552 Waiting time before the next recovery attempt after previous unsuccessful
1553 recovery attempt. Default value is 5 seconds.
1555 @item recovery_wait_streamtime @var{bool}
1556 If set to 0 (false), the real time is used when waiting for the recovery
1557 attempt (i.e. the recovery will be attempted after at least
1558 recovery_wait_time seconds).
1559 If set to 1 (true), the time of the processed stream is taken into account
1560 instead (i.e. the recovery will be attempted after at least @var{recovery_wait_time}
1561 seconds of the stream is omitted).
1562 By default, this option is set to 0 (false).
1564 @item recover_any_error @var{bool}
1565 If set to 1 (true), recovery will be attempted regardless of type of the error
1566 causing the failure. By default this option is set to 0 (false) and in case of
1567 certain (usually permanent) errors the recovery is not attempted even when
1568 @var{attempt_recovery} is set to 1.
1570 @item restart_with_keyframe @var{bool}
1571 Specify whether to wait for the keyframe after recovering from
1572 queue overflow or failure. This option is set to 0 (false) by default.
1576 @subsection Examples
1581 Stream something to rtmp server, continue processing the stream at real-time
1582 rate even in case of temporary failure (network outage) and attempt to recover
1583 streaming every second indefinitely.
1585 ffmpeg -re -i ... -c:v libx264 -c:a aac -f fifo -fifo_format flv -map 0:v -map 0:a
1586 -drop_pkts_on_overflow 1 -attempt_recovery 1 -recovery_wait_time 1 rtmp://example.com/live/stream_name
1594 The tee muxer can be used to write the same data to several files or any
1595 other kind of muxer. It can be used, for example, to both stream a video to
1596 the network and save it to disk at the same time.
1598 It is different from specifying several outputs to the @command{ffmpeg}
1599 command-line tool because the audio and video data will be encoded only once
1600 with the tee muxer; encoding can be a very expensive process. It is not
1601 useful when using the libavformat API directly because it is then possible
1602 to feed the same packets to several muxers directly.
1604 The slave outputs are specified in the file name given to the muxer,
1605 separated by '|'. If any of the slave name contains the '|' separator,
1606 leading or trailing spaces or any special character, it must be
1607 escaped (see @ref{quoting_and_escaping,,the "Quoting and escaping"
1608 section in the ffmpeg-utils(1) manual,ffmpeg-utils}).
1610 Muxer options can be specified for each slave by prepending them as a list of
1611 @var{key}=@var{value} pairs separated by ':', between square brackets. If
1612 the options values contain a special character or the ':' separator, they
1613 must be escaped; note that this is a second level escaping.
1615 The following special options are also recognized:
1618 Specify the format name. Useful if it cannot be guessed from the
1621 @item bsfs[/@var{spec}]
1622 Specify a list of bitstream filters to apply to the specified
1625 It is possible to specify to which streams a given bitstream filter
1626 applies, by appending a stream specifier to the option separated by
1627 @code{/}. @var{spec} must be a stream specifier (see @ref{Format
1628 stream specifiers}). If the stream specifier is not specified, the
1629 bitstream filters will be applied to all streams in the output.
1631 Several bitstream filters can be specified, separated by ",".
1634 Select the streams that should be mapped to the slave output,
1635 specified by a stream specifier. If not specified, this defaults to
1636 all the input streams. You may use multiple stream specifiers
1637 separated by commas (@code{,}) e.g.: @code{a:0,v}
1640 Specify behaviour on output failure. This can be set to either @code{abort} (which is
1641 default) or @code{ignore}. @code{abort} will cause whole process to fail in case of failure
1642 on this slave output. @code{ignore} will ignore failure on this output, so other outputs
1643 will continue without being affected.
1646 @subsection Examples
1650 Encode something and both archive it in a WebM file and stream it
1651 as MPEG-TS over UDP (the streams need to be explicitly mapped):
1653 ffmpeg -i ... -c:v libx264 -c:a mp2 -f tee -map 0:v -map 0:a
1654 "archive-20121107.mkv|[f=mpegts]udp://10.0.1.255:1234/"
1658 As above, but continue streaming even if output to local file fails
1659 (for example local drive fills up):
1661 ffmpeg -i ... -c:v libx264 -c:a mp2 -f tee -map 0:v -map 0:a
1662 "[onfail=ignore]archive-20121107.mkv|[f=mpegts]udp://10.0.1.255:1234/"
1666 Use @command{ffmpeg} to encode the input, and send the output
1667 to three different destinations. The @code{dump_extra} bitstream
1668 filter is used to add extradata information to all the output video
1669 keyframes packets, as requested by the MPEG-TS format. The select
1670 option is applied to @file{out.aac} in order to make it contain only
1673 ffmpeg -i ... -map 0 -flags +global_header -c:v libx264 -c:a aac
1674 -f tee "[bsfs/v=dump_extra]out.ts|[movflags=+faststart]out.mp4|[select=a]out.aac"
1678 As below, but select only stream @code{a:1} for the audio output. Note
1679 that a second level escaping must be performed, as ":" is a special
1680 character used to separate options.
1682 ffmpeg -i ... -map 0 -flags +global_header -c:v libx264 -c:a aac
1683 -f tee "[bsfs/v=dump_extra]out.ts|[movflags=+faststart]out.mp4|[select=\'a:1\']out.aac"
1687 Note: some codecs may need different options depending on the output format;
1688 the auto-detection of this can not work with the tee muxer. The main example
1689 is the @option{global_header} flag.
1691 @section webm_dash_manifest
1693 WebM DASH Manifest muxer.
1695 This muxer implements the WebM DASH Manifest specification to generate the DASH
1696 manifest XML. It also supports manifest generation for DASH live streams.
1698 For more information see:
1702 WebM DASH Specification: @url{https://sites.google.com/a/webmproject.org/wiki/adaptive-streaming/webm-dash-specification}
1704 ISO DASH Specification: @url{http://standards.iso.org/ittf/PubliclyAvailableStandards/c065274_ISO_IEC_23009-1_2014.zip}
1709 This muxer supports the following options:
1712 @item adaptation_sets
1713 This option has the following syntax: "id=x,streams=a,b,c id=y,streams=d,e" where x and y are the
1714 unique identifiers of the adaptation sets and a,b,c,d and e are the indices of the corresponding
1715 audio and video streams. Any number of adaptation sets can be added using this option.
1718 Set this to 1 to create a live stream DASH Manifest. Default: 0.
1720 @item chunk_start_index
1721 Start index of the first chunk. This will go in the @samp{startNumber} attribute
1722 of the @samp{SegmentTemplate} element in the manifest. Default: 0.
1724 @item chunk_duration_ms
1725 Duration of each chunk in milliseconds. This will go in the @samp{duration}
1726 attribute of the @samp{SegmentTemplate} element in the manifest. Default: 1000.
1728 @item utc_timing_url
1729 URL of the page that will return the UTC timestamp in ISO format. This will go
1730 in the @samp{value} attribute of the @samp{UTCTiming} element in the manifest.
1733 @item time_shift_buffer_depth
1734 Smallest time (in seconds) shifting buffer for which any Representation is
1735 guaranteed to be available. This will go in the @samp{timeShiftBufferDepth}
1736 attribute of the @samp{MPD} element. Default: 60.
1738 @item minimum_update_period
1739 Minimum update period (in seconds) of the manifest. This will go in the
1740 @samp{minimumUpdatePeriod} attribute of the @samp{MPD} element. Default: 0.
1746 ffmpeg -f webm_dash_manifest -i video1.webm \
1747 -f webm_dash_manifest -i video2.webm \
1748 -f webm_dash_manifest -i audio1.webm \
1749 -f webm_dash_manifest -i audio2.webm \
1750 -map 0 -map 1 -map 2 -map 3 \
1752 -f webm_dash_manifest \
1753 -adaptation_sets "id=0,streams=0,1 id=1,streams=2,3" \
1759 WebM Live Chunk Muxer.
1761 This muxer writes out WebM headers and chunks as separate files which can be
1762 consumed by clients that support WebM Live streams via DASH.
1766 This muxer supports the following options:
1769 @item chunk_start_index
1770 Index of the first chunk (defaults to 0).
1773 Filename of the header where the initialization data will be written.
1775 @item audio_chunk_duration
1776 Duration of each audio chunk in milliseconds (defaults to 5000).
1781 ffmpeg -f v4l2 -i /dev/video0 \
1785 -s 640x360 -keyint_min 30 -g 30 \
1787 -header webm_live_video_360.hdr \
1788 -chunk_start_index 1 \
1789 webm_live_video_360_%d.chk \
1794 -header webm_live_audio_128.hdr \
1795 -chunk_start_index 1 \
1796 -audio_chunk_duration 1000 \
1797 webm_live_audio_128_%d.chk