1\input texinfo @c -*- texinfo -*- 2@documentencoding UTF-8 3 4@settitle ffmpeg Documentation 5@titlepage 6@center @titlefont{ffmpeg Documentation} 7@end titlepage 8 9@top 10 11@contents 12 13@chapter Synopsis 14 15ffmpeg [@var{global_options}] @{[@var{input_file_options}] -i @file{input_url}@} ... @{[@var{output_file_options}] @file{output_url}@} ... 16 17@chapter Description 18@c man begin DESCRIPTION 19 20@command{ffmpeg} is a very fast video and audio converter that can also grab from 21a live audio/video source. It can also convert between arbitrary sample 22rates and resize video on the fly with a high quality polyphase filter. 23 24@command{ffmpeg} reads from an arbitrary number of input "files" (which can be regular 25files, pipes, network streams, grabbing devices, etc.), specified by the 26@code{-i} option, and writes to an arbitrary number of output "files", which are 27specified by a plain output url. Anything found on the command line which 28cannot be interpreted as an option is considered to be an output url. 29 30Each input or output url can, in principle, contain any number of streams of 31different types (video/audio/subtitle/attachment/data). The allowed number and/or 32types of streams may be limited by the container format. Selecting which 33streams from which inputs will go into which output is either done automatically 34or with the @code{-map} option (see the Stream selection chapter). 35 36To refer to input files in options, you must use their indices (0-based). E.g. 37the first input file is @code{0}, the second is @code{1}, etc. Similarly, streams 38within a file are referred to by their indices. E.g. @code{2:3} refers to the 39fourth stream in the third input file. Also see the Stream specifiers chapter. 40 41As a general rule, options are applied to the next specified 42file. Therefore, order is important, and you can have the same 43option on the command line multiple times. Each occurrence is 44then applied to the next input or output file. 45Exceptions from this rule are the global options (e.g. verbosity level), 46which should be specified first. 47 48Do not mix input and output files -- first specify all input files, then all 49output files. Also do not mix options which belong to different files. All 50options apply ONLY to the next input or output file and are reset between files. 51 52@itemize 53@item 54To set the video bitrate of the output file to 64 kbit/s: 55@example 56ffmpeg -i input.avi -b:v 64k -bufsize 64k output.avi 57@end example 58 59@item 60To force the frame rate of the output file to 24 fps: 61@example 62ffmpeg -i input.avi -r 24 output.avi 63@end example 64 65@item 66To force the frame rate of the input file (valid for raw formats only) 67to 1 fps and the frame rate of the output file to 24 fps: 68@example 69ffmpeg -r 1 -i input.m2v -r 24 output.avi 70@end example 71@end itemize 72 73The format option may be needed for raw input files. 74 75@c man end DESCRIPTION 76 77@chapter Detailed description 78@c man begin DETAILED DESCRIPTION 79 80The transcoding process in @command{ffmpeg} for each output can be described by 81the following diagram: 82 83@verbatim 84 _______ ______________ 85| | | | 86| input | demuxer | encoded data | decoder 87| file | ---------> | packets | -----+ 88|_______| |______________| | 89 v 90 _________ 91 | | 92 | decoded | 93 | frames | 94 |_________| 95 ________ ______________ | 96| | | | | 97| output | <-------- | encoded data | <----+ 98| file | muxer | packets | encoder 99|________| |______________| 100 101 102@end verbatim 103 104@command{ffmpeg} calls the libavformat library (containing demuxers) to read 105input files and get packets containing encoded data from them. When there are 106multiple input files, @command{ffmpeg} tries to keep them synchronized by 107tracking lowest timestamp on any active input stream. 108 109Encoded packets are then passed to the decoder (unless streamcopy is selected 110for the stream, see further for a description). The decoder produces 111uncompressed frames (raw video/PCM audio/...) which can be processed further by 112filtering (see next section). After filtering, the frames are passed to the 113encoder, which encodes them and outputs encoded packets. Finally those are 114passed to the muxer, which writes the encoded packets to the output file. 115 116@section Filtering 117Before encoding, @command{ffmpeg} can process raw audio and video frames using 118filters from the libavfilter library. Several chained filters form a filter 119graph. @command{ffmpeg} distinguishes between two types of filtergraphs: 120simple and complex. 121 122@subsection Simple filtergraphs 123Simple filtergraphs are those that have exactly one input and output, both of 124the same type. In the above diagram they can be represented by simply inserting 125an additional step between decoding and encoding: 126 127@verbatim 128 _________ ______________ 129| | | | 130| decoded | | encoded data | 131| frames |\ _ | packets | 132|_________| \ /||______________| 133 \ __________ / 134 simple _\|| | / encoder 135 filtergraph | filtered |/ 136 | frames | 137 |__________| 138 139@end verbatim 140 141Simple filtergraphs are configured with the per-stream @option{-filter} option 142(with @option{-vf} and @option{-af} aliases for video and audio respectively). 143A simple filtergraph for video can look for example like this: 144 145@verbatim 146 _______ _____________ _______ ________ 147| | | | | | | | 148| input | ---> | deinterlace | ---> | scale | ---> | output | 149|_______| |_____________| |_______| |________| 150 151@end verbatim 152 153Note that some filters change frame properties but not frame contents. E.g. the 154@code{fps} filter in the example above changes number of frames, but does not 155touch the frame contents. Another example is the @code{setpts} filter, which 156only sets timestamps and otherwise passes the frames unchanged. 157 158@subsection Complex filtergraphs 159Complex filtergraphs are those which cannot be described as simply a linear 160processing chain applied to one stream. This is the case, for example, when the graph has 161more than one input and/or output, or when output stream type is different from 162input. They can be represented with the following diagram: 163 164@verbatim 165 _________ 166| | 167| input 0 |\ __________ 168|_________| \ | | 169 \ _________ /| output 0 | 170 \ | | / |__________| 171 _________ \| complex | / 172| | | |/ 173| input 1 |---->| filter |\ 174|_________| | | \ __________ 175 /| graph | \ | | 176 / | | \| output 1 | 177 _________ / |_________| |__________| 178| | / 179| input 2 |/ 180|_________| 181 182@end verbatim 183 184Complex filtergraphs are configured with the @option{-filter_complex} option. 185Note that this option is global, since a complex filtergraph, by its nature, 186cannot be unambiguously associated with a single stream or file. 187 188The @option{-lavfi} option is equivalent to @option{-filter_complex}. 189 190A trivial example of a complex filtergraph is the @code{overlay} filter, which 191has two video inputs and one video output, containing one video overlaid on top 192of the other. Its audio counterpart is the @code{amix} filter. 193 194@section Stream copy 195Stream copy is a mode selected by supplying the @code{copy} parameter to the 196@option{-codec} option. It makes @command{ffmpeg} omit the decoding and encoding 197step for the specified stream, so it does only demuxing and muxing. It is useful 198for changing the container format or modifying container-level metadata. The 199diagram above will, in this case, simplify to this: 200 201@verbatim 202 _______ ______________ ________ 203| | | | | | 204| input | demuxer | encoded data | muxer | output | 205| file | ---------> | packets | -------> | file | 206|_______| |______________| |________| 207 208@end verbatim 209 210Since there is no decoding or encoding, it is very fast and there is no quality 211loss. However, it might not work in some cases because of many factors. Applying 212filters is obviously also impossible, since filters work on uncompressed data. 213 214@c man end DETAILED DESCRIPTION 215 216@chapter Stream selection 217@c man begin STREAM SELECTION 218 219@command{ffmpeg} provides the @code{-map} option for manual control of stream selection in each 220output file. Users can skip @code{-map} and let ffmpeg perform automatic stream selection as 221described below. The @code{-vn / -an / -sn / -dn} options can be used to skip inclusion of 222video, audio, subtitle and data streams respectively, whether manually mapped or automatically 223selected, except for those streams which are outputs of complex filtergraphs. 224 225@section Description 226The sub-sections that follow describe the various rules that are involved in stream selection. 227The examples that follow next show how these rules are applied in practice. 228 229While every effort is made to accurately reflect the behavior of the program, FFmpeg is under 230continuous development and the code may have changed since the time of this writing. 231 232@subsection Automatic stream selection 233 234In the absence of any map options for a particular output file, ffmpeg inspects the output 235format to check which type of streams can be included in it, viz. video, audio and/or 236subtitles. For each acceptable stream type, ffmpeg will pick one stream, when available, 237from among all the inputs. 238 239It will select that stream based upon the following criteria: 240@itemize 241@item 242for video, it is the stream with the highest resolution, 243@item 244for audio, it is the stream with the most channels, 245@item 246for subtitles, it is the first subtitle stream found but there's a caveat. 247The output format's default subtitle encoder can be either text-based or image-based, 248and only a subtitle stream of the same type will be chosen. 249@end itemize 250 251In the case where several streams of the same type rate equally, the stream with the lowest 252index is chosen. 253 254Data or attachment streams are not automatically selected and can only be included 255using @code{-map}. 256@subsection Manual stream selection 257 258When @code{-map} is used, only user-mapped streams are included in that output file, 259with one possible exception for filtergraph outputs described below. 260 261@subsection Complex filtergraphs 262 263If there are any complex filtergraph output streams with unlabeled pads, they will be added 264to the first output file. This will lead to a fatal error if the stream type is not supported 265by the output format. In the absence of the map option, the inclusion of these streams leads 266to the automatic stream selection of their types being skipped. If map options are present, 267these filtergraph streams are included in addition to the mapped streams. 268 269Complex filtergraph output streams with labeled pads must be mapped once and exactly once. 270 271@subsection Stream handling 272 273Stream handling is independent of stream selection, with an exception for subtitles described 274below. Stream handling is set via the @code{-codec} option addressed to streams within a 275specific @emph{output} file. In particular, codec options are applied by ffmpeg after the 276stream selection process and thus do not influence the latter. If no @code{-codec} option is 277specified for a stream type, ffmpeg will select the default encoder registered by the output 278file muxer. 279 280An exception exists for subtitles. If a subtitle encoder is specified for an output file, the 281first subtitle stream found of any type, text or image, will be included. ffmpeg does not validate 282if the specified encoder can convert the selected stream or if the converted stream is acceptable 283within the output format. This applies generally as well: when the user sets an encoder manually, 284the stream selection process cannot check if the encoded stream can be muxed into the output file. 285If it cannot, ffmpeg will abort and @emph{all} output files will fail to be processed. 286 287@section Examples 288 289The following examples illustrate the behavior, quirks and limitations of ffmpeg's stream 290selection methods. 291 292They assume the following three input files. 293 294@verbatim 295 296input file 'A.avi' 297 stream 0: video 640x360 298 stream 1: audio 2 channels 299 300input file 'B.mp4' 301 stream 0: video 1920x1080 302 stream 1: audio 2 channels 303 stream 2: subtitles (text) 304 stream 3: audio 5.1 channels 305 stream 4: subtitles (text) 306 307input file 'C.mkv' 308 stream 0: video 1280x720 309 stream 1: audio 2 channels 310 stream 2: subtitles (image) 311@end verbatim 312 313@subsubheading Example: automatic stream selection 314@example 315ffmpeg -i A.avi -i B.mp4 out1.mkv out2.wav -map 1:a -c:a copy out3.mov 316@end example 317There are three output files specified, and for the first two, no @code{-map} options 318are set, so ffmpeg will select streams for these two files automatically. 319 320@file{out1.mkv} is a Matroska container file and accepts video, audio and subtitle streams, 321so ffmpeg will try to select one of each type.@* 322For video, it will select @code{stream 0} from @file{B.mp4}, which has the highest 323resolution among all the input video streams.@* 324For audio, it will select @code{stream 3} from @file{B.mp4}, since it has the greatest 325number of channels.@* 326For subtitles, it will select @code{stream 2} from @file{B.mp4}, which is the first subtitle 327stream from among @file{A.avi} and @file{B.mp4}. 328 329@file{out2.wav} accepts only audio streams, so only @code{stream 3} from @file{B.mp4} is 330selected. 331 332For @file{out3.mov}, since a @code{-map} option is set, no automatic stream selection will 333occur. The @code{-map 1:a} option will select all audio streams from the second input 334@file{B.mp4}. No other streams will be included in this output file. 335 336For the first two outputs, all included streams will be transcoded. The encoders chosen will 337be the default ones registered by each output format, which may not match the codec of the 338selected input streams. 339 340For the third output, codec option for audio streams has been set 341to @code{copy}, so no decoding-filtering-encoding operations will occur, or @emph{can} occur. 342Packets of selected streams shall be conveyed from the input file and muxed within the output 343file. 344 345@subsubheading Example: automatic subtitles selection 346@example 347ffmpeg -i C.mkv out1.mkv -c:s dvdsub -an out2.mkv 348@end example 349Although @file{out1.mkv} is a Matroska container file which accepts subtitle streams, only a 350video and audio stream shall be selected. The subtitle stream of @file{C.mkv} is image-based 351and the default subtitle encoder of the Matroska muxer is text-based, so a transcode operation 352for the subtitles is expected to fail and hence the stream isn't selected. However, in 353@file{out2.mkv}, a subtitle encoder is specified in the command and so, the subtitle stream is 354selected, in addition to the video stream. The presence of @code{-an} disables audio stream 355selection for @file{out2.mkv}. 356 357@subsubheading Example: unlabeled filtergraph outputs 358@example 359ffmpeg -i A.avi -i C.mkv -i B.mp4 -filter_complex "overlay" out1.mp4 out2.srt 360@end example 361A filtergraph is setup here using the @code{-filter_complex} option and consists of a single 362video filter. The @code{overlay} filter requires exactly two video inputs, but none are 363specified, so the first two available video streams are used, those of @file{A.avi} and 364@file{C.mkv}. The output pad of the filter has no label and so is sent to the first output file 365@file{out1.mp4}. Due to this, automatic selection of the video stream is skipped, which would 366have selected the stream in @file{B.mp4}. The audio stream with most channels viz. @code{stream 3} 367in @file{B.mp4}, is chosen automatically. No subtitle stream is chosen however, since the MP4 368format has no default subtitle encoder registered, and the user hasn't specified a subtitle encoder. 369 370The 2nd output file, @file{out2.srt}, only accepts text-based subtitle streams. So, even though 371the first subtitle stream available belongs to @file{C.mkv}, it is image-based and hence skipped. 372The selected stream, @code{stream 2} in @file{B.mp4}, is the first text-based subtitle stream. 373 374@subsubheading Example: labeled filtergraph outputs 375@example 376ffmpeg -i A.avi -i B.mp4 -i C.mkv -filter_complex "[1:v]hue=s=0[outv];overlay;aresample" \ 377 -map '[outv]' -an out1.mp4 \ 378 out2.mkv \ 379 -map '[outv]' -map 1:a:0 out3.mkv 380@end example 381 382The above command will fail, as the output pad labelled @code{[outv]} has been mapped twice. 383None of the output files shall be processed. 384 385@example 386ffmpeg -i A.avi -i B.mp4 -i C.mkv -filter_complex "[1:v]hue=s=0[outv];overlay;aresample" \ 387 -an out1.mp4 \ 388 out2.mkv \ 389 -map 1:a:0 out3.mkv 390@end example 391 392This command above will also fail as the hue filter output has a label, @code{[outv]}, 393and hasn't been mapped anywhere. 394 395The command should be modified as follows, 396@example 397ffmpeg -i A.avi -i B.mp4 -i C.mkv -filter_complex "[1:v]hue=s=0,split=2[outv1][outv2];overlay;aresample" \ 398 -map '[outv1]' -an out1.mp4 \ 399 out2.mkv \ 400 -map '[outv2]' -map 1:a:0 out3.mkv 401@end example 402The video stream from @file{B.mp4} is sent to the hue filter, whose output is cloned once using 403the split filter, and both outputs labelled. Then a copy each is mapped to the first and third 404output files. 405 406The overlay filter, requiring two video inputs, uses the first two unused video streams. Those 407are the streams from @file{A.avi} and @file{C.mkv}. The overlay output isn't labelled, so it is 408sent to the first output file @file{out1.mp4}, regardless of the presence of the @code{-map} option. 409 410The aresample filter is sent the first unused audio stream, that of @file{A.avi}. Since this filter 411output is also unlabelled, it too is mapped to the first output file. The presence of @code{-an} 412only suppresses automatic or manual stream selection of audio streams, not outputs sent from 413filtergraphs. Both these mapped streams shall be ordered before the mapped stream in @file{out1.mp4}. 414 415The video, audio and subtitle streams mapped to @code{out2.mkv} are entirely determined by 416automatic stream selection. 417 418@file{out3.mkv} consists of the cloned video output from the hue filter and the first audio 419stream from @file{B.mp4}. 420@* 421 422@c man end STREAM SELECTION 423 424@chapter Options 425@c man begin OPTIONS 426 427@include fftools-common-opts.texi 428 429@section Main options 430 431@table @option 432 433@item -f @var{fmt} (@emph{input/output}) 434Force input or output file format. The format is normally auto detected for input 435files and guessed from the file extension for output files, so this option is not 436needed in most cases. 437 438@item -i @var{url} (@emph{input}) 439input file url 440 441@item -y (@emph{global}) 442Overwrite output files without asking. 443 444@item -n (@emph{global}) 445Do not overwrite output files, and exit immediately if a specified 446output file already exists. 447 448@item -stream_loop @var{number} (@emph{input}) 449Set number of times input stream shall be looped. Loop 0 means no loop, 450loop -1 means infinite loop. 451 452@item -c[:@var{stream_specifier}] @var{codec} (@emph{input/output,per-stream}) 453@itemx -codec[:@var{stream_specifier}] @var{codec} (@emph{input/output,per-stream}) 454Select an encoder (when used before an output file) or a decoder (when used 455before an input file) for one or more streams. @var{codec} is the name of a 456decoder/encoder or a special value @code{copy} (output only) to indicate that 457the stream is not to be re-encoded. 458 459For example 460@example 461ffmpeg -i INPUT -map 0 -c:v libx264 -c:a copy OUTPUT 462@end example 463encodes all video streams with libx264 and copies all audio streams. 464 465For each stream, the last matching @code{c} option is applied, so 466@example 467ffmpeg -i INPUT -map 0 -c copy -c:v:1 libx264 -c:a:137 libvorbis OUTPUT 468@end example 469will copy all the streams except the second video, which will be encoded with 470libx264, and the 138th audio, which will be encoded with libvorbis. 471 472@item -t @var{duration} (@emph{input/output}) 473When used as an input option (before @code{-i}), limit the @var{duration} of 474data read from the input file. 475 476When used as an output option (before an output url), stop writing the 477output after its duration reaches @var{duration}. 478 479@var{duration} must be a time duration specification, 480see @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}. 481 482-to and -t are mutually exclusive and -t has priority. 483 484@item -to @var{position} (@emph{input/output}) 485Stop writing the output or reading the input at @var{position}. 486@var{position} must be a time duration specification, 487see @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}. 488 489-to and -t are mutually exclusive and -t has priority. 490 491@item -fs @var{limit_size} (@emph{output}) 492Set the file size limit, expressed in bytes. No further chunk of bytes is written 493after the limit is exceeded. The size of the output file is slightly more than the 494requested file size. 495 496@item -ss @var{position} (@emph{input/output}) 497When used as an input option (before @code{-i}), seeks in this input file to 498@var{position}. Note that in most formats it is not possible to seek exactly, 499so @command{ffmpeg} will seek to the closest seek point before @var{position}. 500When transcoding and @option{-accurate_seek} is enabled (the default), this 501extra segment between the seek point and @var{position} will be decoded and 502discarded. When doing stream copy or when @option{-noaccurate_seek} is used, it 503will be preserved. 504 505When used as an output option (before an output url), decodes but discards 506input until the timestamps reach @var{position}. 507 508@var{position} must be a time duration specification, 509see @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}. 510 511@item -sseof @var{position} (@emph{input}) 512 513Like the @code{-ss} option but relative to the "end of file". That is negative 514values are earlier in the file, 0 is at EOF. 515 516@item -itsoffset @var{offset} (@emph{input}) 517Set the input time offset. 518 519@var{offset} must be a time duration specification, 520see @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}. 521 522The offset is added to the timestamps of the input files. Specifying 523a positive offset means that the corresponding streams are delayed by 524the time duration specified in @var{offset}. 525 526@item -itsscale @var{scale} (@emph{input,per-stream}) 527Rescale input timestamps. @var{scale} should be a floating point number. 528 529@item -timestamp @var{date} (@emph{output}) 530Set the recording timestamp in the container. 531 532@var{date} must be a date specification, 533see @ref{date syntax,,the Date section in the ffmpeg-utils(1) manual,ffmpeg-utils}. 534 535@item -metadata[:metadata_specifier] @var{key}=@var{value} (@emph{output,per-metadata}) 536Set a metadata key/value pair. 537 538An optional @var{metadata_specifier} may be given to set metadata 539on streams, chapters or programs. See @code{-map_metadata} 540documentation for details. 541 542This option overrides metadata set with @code{-map_metadata}. It is 543also possible to delete metadata by using an empty value. 544 545For example, for setting the title in the output file: 546@example 547ffmpeg -i in.avi -metadata title="my title" out.flv 548@end example 549 550To set the language of the first audio stream: 551@example 552ffmpeg -i INPUT -metadata:s:a:0 language=eng OUTPUT 553@end example 554 555@item -disposition[:stream_specifier] @var{value} (@emph{output,per-stream}) 556Sets the disposition for a stream. 557 558This option overrides the disposition copied from the input stream. It is also 559possible to delete the disposition by setting it to 0. 560 561The following dispositions are recognized: 562@table @option 563@item default 564@item dub 565@item original 566@item comment 567@item lyrics 568@item karaoke 569@item forced 570@item hearing_impaired 571@item visual_impaired 572@item clean_effects 573@item attached_pic 574@item captions 575@item descriptions 576@item dependent 577@item metadata 578@end table 579 580For example, to make the second audio stream the default stream: 581@example 582ffmpeg -i in.mkv -c copy -disposition:a:1 default out.mkv 583@end example 584 585To make the second subtitle stream the default stream and remove the default 586disposition from the first subtitle stream: 587@example 588ffmpeg -i in.mkv -c copy -disposition:s:0 0 -disposition:s:1 default out.mkv 589@end example 590 591To add an embedded cover/thumbnail: 592@example 593ffmpeg -i in.mp4 -i IMAGE -map 0 -map 1 -c copy -c:v:1 png -disposition:v:1 attached_pic out.mp4 594@end example 595 596Not all muxers support embedded thumbnails, and those who do, only support a few formats, like JPEG or PNG. 597 598@item -program [title=@var{title}:][program_num=@var{program_num}:]st=@var{stream}[:st=@var{stream}...] (@emph{output}) 599 600Creates a program with the specified @var{title}, @var{program_num} and adds the specified 601@var{stream}(s) to it. 602 603@item -target @var{type} (@emph{output}) 604Specify target file type (@code{vcd}, @code{svcd}, @code{dvd}, @code{dv}, 605@code{dv50}). @var{type} may be prefixed with @code{pal-}, @code{ntsc-} or 606@code{film-} to use the corresponding standard. All the format options 607(bitrate, codecs, buffer sizes) are then set automatically. You can just type: 608 609@example 610ffmpeg -i myfile.avi -target vcd /tmp/vcd.mpg 611@end example 612 613Nevertheless you can specify additional options as long as you know 614they do not conflict with the standard, as in: 615 616@example 617ffmpeg -i myfile.avi -target vcd -bf 2 /tmp/vcd.mpg 618@end example 619 620@item -dn (@emph{input/output}) 621As an input option, blocks all data streams of a file from being filtered or 622being automatically selected or mapped for any output. See @code{-discard} 623option to disable streams individually. 624 625As an output option, disables data recording i.e. automatic selection or 626mapping of any data stream. For full manual control see the @code{-map} 627option. 628 629@item -dframes @var{number} (@emph{output}) 630Set the number of data frames to output. This is an obsolete alias for 631@code{-frames:d}, which you should use instead. 632 633@item -frames[:@var{stream_specifier}] @var{framecount} (@emph{output,per-stream}) 634Stop writing to the stream after @var{framecount} frames. 635 636@item -q[:@var{stream_specifier}] @var{q} (@emph{output,per-stream}) 637@itemx -qscale[:@var{stream_specifier}] @var{q} (@emph{output,per-stream}) 638Use fixed quality scale (VBR). The meaning of @var{q}/@var{qscale} is 639codec-dependent. 640If @var{qscale} is used without a @var{stream_specifier} then it applies only 641to the video stream, this is to maintain compatibility with previous behavior 642and as specifying the same codec specific value to 2 different codecs that is 643audio and video generally is not what is intended when no stream_specifier is 644used. 645 646@anchor{filter_option} 647@item -filter[:@var{stream_specifier}] @var{filtergraph} (@emph{output,per-stream}) 648Create the filtergraph specified by @var{filtergraph} and use it to 649filter the stream. 650 651@var{filtergraph} is a description of the filtergraph to apply to 652the stream, and must have a single input and a single output of the 653same type of the stream. In the filtergraph, the input is associated 654to the label @code{in}, and the output to the label @code{out}. See 655the ffmpeg-filters manual for more information about the filtergraph 656syntax. 657 658See the @ref{filter_complex_option,,-filter_complex option} if you 659want to create filtergraphs with multiple inputs and/or outputs. 660 661@item -filter_script[:@var{stream_specifier}] @var{filename} (@emph{output,per-stream}) 662This option is similar to @option{-filter}, the only difference is that its 663argument is the name of the file from which a filtergraph description is to be 664read. 665 666@item -filter_threads @var{nb_threads} (@emph{global}) 667Defines how many threads are used to process a filter pipeline. Each pipeline 668will produce a thread pool with this many threads available for parallel processing. 669The default is the number of available CPUs. 670 671@item -pre[:@var{stream_specifier}] @var{preset_name} (@emph{output,per-stream}) 672Specify the preset for matching stream(s). 673 674@item -stats (@emph{global}) 675Print encoding progress/statistics. It is on by default, to explicitly 676disable it you need to specify @code{-nostats}. 677 678@item -progress @var{url} (@emph{global}) 679Send program-friendly progress information to @var{url}. 680 681Progress information is written approximately every second and at the end of 682the encoding process. It is made of "@var{key}=@var{value}" lines. @var{key} 683consists of only alphanumeric characters. The last key of a sequence of 684progress information is always "progress". 685 686@anchor{stdin option} 687@item -stdin 688Enable interaction on standard input. On by default unless standard input is 689used as an input. To explicitly disable interaction you need to specify 690@code{-nostdin}. 691 692Disabling interaction on standard input is useful, for example, if 693ffmpeg is in the background process group. Roughly the same result can 694be achieved with @code{ffmpeg ... < /dev/null} but it requires a 695shell. 696 697@item -debug_ts (@emph{global}) 698Print timestamp information. It is off by default. This option is 699mostly useful for testing and debugging purposes, and the output 700format may change from one version to another, so it should not be 701employed by portable scripts. 702 703See also the option @code{-fdebug ts}. 704 705@item -attach @var{filename} (@emph{output}) 706Add an attachment to the output file. This is supported by a few formats 707like Matroska for e.g. fonts used in rendering subtitles. Attachments 708are implemented as a specific type of stream, so this option will add 709a new stream to the file. It is then possible to use per-stream options 710on this stream in the usual way. Attachment streams created with this 711option will be created after all the other streams (i.e. those created 712with @code{-map} or automatic mappings). 713 714Note that for Matroska you also have to set the mimetype metadata tag: 715@example 716ffmpeg -i INPUT -attach DejaVuSans.ttf -metadata:s:2 mimetype=application/x-truetype-font out.mkv 717@end example 718(assuming that the attachment stream will be third in the output file). 719 720@item -dump_attachment[:@var{stream_specifier}] @var{filename} (@emph{input,per-stream}) 721Extract the matching attachment stream into a file named @var{filename}. If 722@var{filename} is empty, then the value of the @code{filename} metadata tag 723will be used. 724 725E.g. to extract the first attachment to a file named 'out.ttf': 726@example 727ffmpeg -dump_attachment:t:0 out.ttf -i INPUT 728@end example 729To extract all attachments to files determined by the @code{filename} tag: 730@example 731ffmpeg -dump_attachment:t "" -i INPUT 732@end example 733 734Technical note -- attachments are implemented as codec extradata, so this 735option can actually be used to extract extradata from any stream, not just 736attachments. 737 738@item -noautorotate 739Disable automatically rotating video based on file metadata. 740 741@end table 742 743@section Video Options 744 745@table @option 746@item -vframes @var{number} (@emph{output}) 747Set the number of video frames to output. This is an obsolete alias for 748@code{-frames:v}, which you should use instead. 749@item -r[:@var{stream_specifier}] @var{fps} (@emph{input/output,per-stream}) 750Set frame rate (Hz value, fraction or abbreviation). 751 752As an input option, ignore any timestamps stored in the file and instead 753generate timestamps assuming constant frame rate @var{fps}. 754This is not the same as the @option{-framerate} option used for some input formats 755like image2 or v4l2 (it used to be the same in older versions of FFmpeg). 756If in doubt use @option{-framerate} instead of the input option @option{-r}. 757 758As an output option, duplicate or drop input frames to achieve constant output 759frame rate @var{fps}. 760 761@item -s[:@var{stream_specifier}] @var{size} (@emph{input/output,per-stream}) 762Set frame size. 763 764As an input option, this is a shortcut for the @option{video_size} private 765option, recognized by some demuxers for which the frame size is either not 766stored in the file or is configurable -- e.g. raw video or video grabbers. 767 768As an output option, this inserts the @code{scale} video filter to the 769@emph{end} of the corresponding filtergraph. Please use the @code{scale} filter 770directly to insert it at the beginning or some other place. 771 772The format is @samp{wxh} (default - same as source). 773 774@item -aspect[:@var{stream_specifier}] @var{aspect} (@emph{output,per-stream}) 775Set the video display aspect ratio specified by @var{aspect}. 776 777@var{aspect} can be a floating point number string, or a string of the 778form @var{num}:@var{den}, where @var{num} and @var{den} are the 779numerator and denominator of the aspect ratio. For example "4:3", 780"16:9", "1.3333", and "1.7777" are valid argument values. 781 782If used together with @option{-vcodec copy}, it will affect the aspect ratio 783stored at container level, but not the aspect ratio stored in encoded 784frames, if it exists. 785 786@item -vn (@emph{input/output}) 787As an input option, blocks all video streams of a file from being filtered or 788being automatically selected or mapped for any output. See @code{-discard} 789option to disable streams individually. 790 791As an output option, disables video recording i.e. automatic selection or 792mapping of any video stream. For full manual control see the @code{-map} 793option. 794 795@item -vcodec @var{codec} (@emph{output}) 796Set the video codec. This is an alias for @code{-codec:v}. 797 798@item -pass[:@var{stream_specifier}] @var{n} (@emph{output,per-stream}) 799Select the pass number (1 or 2). It is used to do two-pass 800video encoding. The statistics of the video are recorded in the first 801pass into a log file (see also the option -passlogfile), 802and in the second pass that log file is used to generate the video 803at the exact requested bitrate. 804On pass 1, you may just deactivate audio and set output to null, 805examples for Windows and Unix: 806@example 807ffmpeg -i foo.mov -c:v libxvid -pass 1 -an -f rawvideo -y NUL 808ffmpeg -i foo.mov -c:v libxvid -pass 1 -an -f rawvideo -y /dev/null 809@end example 810 811@item -passlogfile[:@var{stream_specifier}] @var{prefix} (@emph{output,per-stream}) 812Set two-pass log file name prefix to @var{prefix}, the default file name 813prefix is ``ffmpeg2pass''. The complete file name will be 814@file{PREFIX-N.log}, where N is a number specific to the output 815stream 816 817@item -vf @var{filtergraph} (@emph{output}) 818Create the filtergraph specified by @var{filtergraph} and use it to 819filter the stream. 820 821This is an alias for @code{-filter:v}, see the @ref{filter_option,,-filter option}. 822@end table 823 824@section Advanced Video options 825 826@table @option 827@item -pix_fmt[:@var{stream_specifier}] @var{format} (@emph{input/output,per-stream}) 828Set pixel format. Use @code{-pix_fmts} to show all the supported 829pixel formats. 830If the selected pixel format can not be selected, ffmpeg will print a 831warning and select the best pixel format supported by the encoder. 832If @var{pix_fmt} is prefixed by a @code{+}, ffmpeg will exit with an error 833if the requested pixel format can not be selected, and automatic conversions 834inside filtergraphs are disabled. 835If @var{pix_fmt} is a single @code{+}, ffmpeg selects the same pixel format 836as the input (or graph output) and automatic conversions are disabled. 837 838@item -sws_flags @var{flags} (@emph{input/output}) 839Set SwScaler flags. 840 841@item -rc_override[:@var{stream_specifier}] @var{override} (@emph{output,per-stream}) 842Rate control override for specific intervals, formatted as "int,int,int" 843list separated with slashes. Two first values are the beginning and 844end frame numbers, last one is quantizer to use if positive, or quality 845factor if negative. 846 847@item -ilme 848Force interlacing support in encoder (MPEG-2 and MPEG-4 only). 849Use this option if your input file is interlaced and you want 850to keep the interlaced format for minimum losses. 851The alternative is to deinterlace the input stream with 852@option{-deinterlace}, but deinterlacing introduces losses. 853@item -psnr 854Calculate PSNR of compressed frames. 855@item -vstats 856Dump video coding statistics to @file{vstats_HHMMSS.log}. 857@item -vstats_file @var{file} 858Dump video coding statistics to @var{file}. 859@item -vstats_version @var{file} 860Specifies which version of the vstats format to use. Default is 2. 861 862version = 1 : 863 864@code{frame= %5d q= %2.1f PSNR= %6.2f f_size= %6d s_size= %8.0fkB time= %0.3f br= %7.1fkbits/s avg_br= %7.1fkbits/s} 865 866version > 1: 867 868@code{out= %2d st= %2d frame= %5d q= %2.1f PSNR= %6.2f f_size= %6d s_size= %8.0fkB time= %0.3f br= %7.1fkbits/s avg_br= %7.1fkbits/s} 869@item -top[:@var{stream_specifier}] @var{n} (@emph{output,per-stream}) 870top=1/bottom=0/auto=-1 field first 871@item -dc @var{precision} 872Intra_dc_precision. 873@item -vtag @var{fourcc/tag} (@emph{output}) 874Force video tag/fourcc. This is an alias for @code{-tag:v}. 875@item -qphist (@emph{global}) 876Show QP histogram 877@item -vbsf @var{bitstream_filter} 878Deprecated see -bsf 879 880@item -force_key_frames[:@var{stream_specifier}] @var{time}[,@var{time}...] (@emph{output,per-stream}) 881@item -force_key_frames[:@var{stream_specifier}] expr:@var{expr} (@emph{output,per-stream}) 882@item -force_key_frames[:@var{stream_specifier}] source (@emph{output,per-stream}) 883 884@var{force_key_frames} can take arguments of the following form: 885 886@table @option 887 888@item @var{time}[,@var{time}...] 889If the argument consists of timestamps, ffmpeg will round the specified times to the nearest 890output timestamp as per the encoder time base and force a keyframe at the first frame having 891timestamp equal or greater than the computed timestamp. Note that if the encoder time base is too 892coarse, then the keyframes may be forced on frames with timestamps lower than the specified time. 893The default encoder time base is the inverse of the output framerate but may be set otherwise 894via @code{-enc_time_base}. 895 896If one of the times is "@code{chapters}[@var{delta}]", it is expanded into 897the time of the beginning of all chapters in the file, shifted by 898@var{delta}, expressed as a time in seconds. 899This option can be useful to ensure that a seek point is present at a 900chapter mark or any other designated place in the output file. 901 902For example, to insert a key frame at 5 minutes, plus key frames 0.1 second 903before the beginning of every chapter: 904@example 905-force_key_frames 0:05:00,chapters-0.1 906@end example 907 908@item expr:@var{expr} 909If the argument is prefixed with @code{expr:}, the string @var{expr} 910is interpreted like an expression and is evaluated for each frame. A 911key frame is forced in case the evaluation is non-zero. 912 913The expression in @var{expr} can contain the following constants: 914@table @option 915@item n 916the number of current processed frame, starting from 0 917@item n_forced 918the number of forced frames 919@item prev_forced_n 920the number of the previous forced frame, it is @code{NAN} when no 921keyframe was forced yet 922@item prev_forced_t 923the time of the previous forced frame, it is @code{NAN} when no 924keyframe was forced yet 925@item t 926the time of the current processed frame 927@end table 928 929For example to force a key frame every 5 seconds, you can specify: 930@example 931-force_key_frames expr:gte(t,n_forced*5) 932@end example 933 934To force a key frame 5 seconds after the time of the last forced one, 935starting from second 13: 936@example 937-force_key_frames expr:if(isnan(prev_forced_t),gte(t,13),gte(t,prev_forced_t+5)) 938@end example 939 940@item source 941If the argument is @code{source}, ffmpeg will force a key frame if 942the current frame being encoded is marked as a key frame in its source. 943 944@end table 945 946Note that forcing too many keyframes is very harmful for the lookahead 947algorithms of certain encoders: using fixed-GOP options or similar 948would be more efficient. 949 950@item -copyinkf[:@var{stream_specifier}] (@emph{output,per-stream}) 951When doing stream copy, copy also non-key frames found at the 952beginning. 953 954@item -init_hw_device @var{type}[=@var{name}][:@var{device}[,@var{key=value}...]] 955Initialise a new hardware device of type @var{type} called @var{name}, using the 956given device parameters. 957If no name is specified it will receive a default name of the form "@var{type}%d". 958 959The meaning of @var{device} and the following arguments depends on the 960device type: 961@table @option 962 963@item cuda 964@var{device} is the number of the CUDA device. 965 966@item dxva2 967@var{device} is the number of the Direct3D 9 display adapter. 968 969@item vaapi 970@var{device} is either an X11 display name or a DRM render node. 971If not specified, it will attempt to open the default X11 display (@emph{$DISPLAY}) 972and then the first DRM render node (@emph{/dev/dri/renderD128}). 973 974@item vdpau 975@var{device} is an X11 display name. 976If not specified, it will attempt to open the default X11 display (@emph{$DISPLAY}). 977 978@item qsv 979@var{device} selects a value in @samp{MFX_IMPL_*}. Allowed values are: 980@table @option 981@item auto 982@item sw 983@item hw 984@item auto_any 985@item hw_any 986@item hw2 987@item hw3 988@item hw4 989@end table 990If not specified, @samp{auto_any} is used. 991(Note that it may be easier to achieve the desired result for QSV by creating the 992platform-appropriate subdevice (@samp{dxva2} or @samp{vaapi}) and then deriving a 993QSV device from that.) 994 995@item opencl 996@var{device} selects the platform and device as @emph{platform_index.device_index}. 997 998The set of devices can also be filtered using the key-value pairs to find only 999devices matching particular platform or device strings. 1000 1001The strings usable as filters are: 1002@table @option 1003@item platform_profile 1004@item platform_version 1005@item platform_name 1006@item platform_vendor 1007@item platform_extensions 1008@item device_name 1009@item device_vendor 1010@item driver_version 1011@item device_version 1012@item device_profile 1013@item device_extensions 1014@item device_type 1015@end table 1016 1017The indices and filters must together uniquely select a device. 1018 1019Examples: 1020@table @emph 1021@item -init_hw_device opencl:0.1 1022Choose the second device on the first platform. 1023 1024@item -init_hw_device opencl:,device_name=Foo9000 1025Choose the device with a name containing the string @emph{Foo9000}. 1026 1027@item -init_hw_device opencl:1,device_type=gpu,device_extensions=cl_khr_fp16 1028Choose the GPU device on the second platform supporting the @emph{cl_khr_fp16} 1029extension. 1030@end table 1031 1032@item vulkan 1033If @var{device} is an integer, it selects the device by its index in a 1034system-dependent list of devices. If @var{device} is any other string, it 1035selects the first device with a name containing that string as a substring. 1036 1037The following options are recognized: 1038@table @option 1039@item debug 1040If set to 1, enables the validation layer, if installed. 1041@item linear_images 1042If set to 1, images allocated by the hwcontext will be linear and locally mappable. 1043@item instance_extensions 1044A plus separated list of additional instance extensions to enable. 1045@item device_extensions 1046A plus separated list of additional device extensions to enable. 1047@end table 1048 1049Examples: 1050@table @emph 1051@item -init_hw_device vulkan:1 1052Choose the second device on the system. 1053 1054@item -init_hw_device vulkan:RADV 1055Choose the first device with a name containing the string @emph{RADV}. 1056 1057@item -init_hw_device vulkan:0,instance_extensions=VK_KHR_wayland_surface+VK_KHR_xcb_surface 1058Choose the first device and enable the Wayland and XCB instance extensions. 1059@end table 1060 1061@end table 1062 1063@item -init_hw_device @var{type}[=@var{name}]@@@var{source} 1064Initialise a new hardware device of type @var{type} called @var{name}, 1065deriving it from the existing device with the name @var{source}. 1066 1067@item -init_hw_device list 1068List all hardware device types supported in this build of ffmpeg. 1069 1070@item -filter_hw_device @var{name} 1071Pass the hardware device called @var{name} to all filters in any filter graph. 1072This can be used to set the device to upload to with the @code{hwupload} filter, 1073or the device to map to with the @code{hwmap} filter. Other filters may also 1074make use of this parameter when they require a hardware device. Note that this 1075is typically only required when the input is not already in hardware frames - 1076when it is, filters will derive the device they require from the context of the 1077frames they receive as input. 1078 1079This is a global setting, so all filters will receive the same device. 1080 1081@item -hwaccel[:@var{stream_specifier}] @var{hwaccel} (@emph{input,per-stream}) 1082Use hardware acceleration to decode the matching stream(s). The allowed values 1083of @var{hwaccel} are: 1084@table @option 1085@item none 1086Do not use any hardware acceleration (the default). 1087 1088@item auto 1089Automatically select the hardware acceleration method. 1090 1091@item vdpau 1092Use VDPAU (Video Decode and Presentation API for Unix) hardware acceleration. 1093 1094@item dxva2 1095Use DXVA2 (DirectX Video Acceleration) hardware acceleration. 1096 1097@item vaapi 1098Use VAAPI (Video Acceleration API) hardware acceleration. 1099 1100@item qsv 1101Use the Intel QuickSync Video acceleration for video transcoding. 1102 1103Unlike most other values, this option does not enable accelerated decoding (that 1104is used automatically whenever a qsv decoder is selected), but accelerated 1105transcoding, without copying the frames into the system memory. 1106 1107For it to work, both the decoder and the encoder must support QSV acceleration 1108and no filters must be used. 1109@end table 1110 1111This option has no effect if the selected hwaccel is not available or not 1112supported by the chosen decoder. 1113 1114Note that most acceleration methods are intended for playback and will not be 1115faster than software decoding on modern CPUs. Additionally, @command{ffmpeg} 1116will usually need to copy the decoded frames from the GPU memory into the system 1117memory, resulting in further performance loss. This option is thus mainly 1118useful for testing. 1119 1120@item -hwaccel_device[:@var{stream_specifier}] @var{hwaccel_device} (@emph{input,per-stream}) 1121Select a device to use for hardware acceleration. 1122 1123This option only makes sense when the @option{-hwaccel} option is also specified. 1124It can either refer to an existing device created with @option{-init_hw_device} 1125by name, or it can create a new device as if 1126@samp{-init_hw_device} @var{type}:@var{hwaccel_device} 1127were called immediately before. 1128 1129@item -hwaccels 1130List all hardware acceleration methods supported in this build of ffmpeg. 1131 1132@end table 1133 1134@section Audio Options 1135 1136@table @option 1137@item -aframes @var{number} (@emph{output}) 1138Set the number of audio frames to output. This is an obsolete alias for 1139@code{-frames:a}, which you should use instead. 1140@item -ar[:@var{stream_specifier}] @var{freq} (@emph{input/output,per-stream}) 1141Set the audio sampling frequency. For output streams it is set by 1142default to the frequency of the corresponding input stream. For input 1143streams this option only makes sense for audio grabbing devices and raw 1144demuxers and is mapped to the corresponding demuxer options. 1145@item -aq @var{q} (@emph{output}) 1146Set the audio quality (codec-specific, VBR). This is an alias for -q:a. 1147@item -ac[:@var{stream_specifier}] @var{channels} (@emph{input/output,per-stream}) 1148Set the number of audio channels. For output streams it is set by 1149default to the number of input audio channels. For input streams 1150this option only makes sense for audio grabbing devices and raw demuxers 1151and is mapped to the corresponding demuxer options. 1152@item -an (@emph{input/output}) 1153As an input option, blocks all audio streams of a file from being filtered or 1154being automatically selected or mapped for any output. See @code{-discard} 1155option to disable streams individually. 1156 1157As an output option, disables audio recording i.e. automatic selection or 1158mapping of any audio stream. For full manual control see the @code{-map} 1159option. 1160@item -acodec @var{codec} (@emph{input/output}) 1161Set the audio codec. This is an alias for @code{-codec:a}. 1162@item -sample_fmt[:@var{stream_specifier}] @var{sample_fmt} (@emph{output,per-stream}) 1163Set the audio sample format. Use @code{-sample_fmts} to get a list 1164of supported sample formats. 1165 1166@item -af @var{filtergraph} (@emph{output}) 1167Create the filtergraph specified by @var{filtergraph} and use it to 1168filter the stream. 1169 1170This is an alias for @code{-filter:a}, see the @ref{filter_option,,-filter option}. 1171@end table 1172 1173@section Advanced Audio options 1174 1175@table @option 1176@item -atag @var{fourcc/tag} (@emph{output}) 1177Force audio tag/fourcc. This is an alias for @code{-tag:a}. 1178@item -absf @var{bitstream_filter} 1179Deprecated, see -bsf 1180@item -guess_layout_max @var{channels} (@emph{input,per-stream}) 1181If some input channel layout is not known, try to guess only if it 1182corresponds to at most the specified number of channels. For example, 2 1183tells to @command{ffmpeg} to recognize 1 channel as mono and 2 channels as 1184stereo but not 6 channels as 5.1. The default is to always try to guess. Use 11850 to disable all guessing. 1186@end table 1187 1188@section Subtitle options 1189 1190@table @option 1191@item -scodec @var{codec} (@emph{input/output}) 1192Set the subtitle codec. This is an alias for @code{-codec:s}. 1193@item -sn (@emph{input/output}) 1194As an input option, blocks all subtitle streams of a file from being filtered or 1195being automatically selected or mapped for any output. See @code{-discard} 1196option to disable streams individually. 1197 1198As an output option, disables subtitle recording i.e. automatic selection or 1199mapping of any subtitle stream. For full manual control see the @code{-map} 1200option. 1201@item -sbsf @var{bitstream_filter} 1202Deprecated, see -bsf 1203@end table 1204 1205@section Advanced Subtitle options 1206 1207@table @option 1208 1209@item -fix_sub_duration 1210Fix subtitles durations. For each subtitle, wait for the next packet in the 1211same stream and adjust the duration of the first to avoid overlap. This is 1212necessary with some subtitles codecs, especially DVB subtitles, because the 1213duration in the original packet is only a rough estimate and the end is 1214actually marked by an empty subtitle frame. Failing to use this option when 1215necessary can result in exaggerated durations or muxing failures due to 1216non-monotonic timestamps. 1217 1218Note that this option will delay the output of all data until the next 1219subtitle packet is decoded: it may increase memory consumption and latency a 1220lot. 1221 1222@item -canvas_size @var{size} 1223Set the size of the canvas used to render subtitles. 1224 1225@end table 1226 1227@section Advanced options 1228 1229@table @option 1230@item -map [-]@var{input_file_id}[:@var{stream_specifier}][?][,@var{sync_file_id}[:@var{stream_specifier}]] | @var{[linklabel]} (@emph{output}) 1231 1232Designate one or more input streams as a source for the output file. Each input 1233stream is identified by the input file index @var{input_file_id} and 1234the input stream index @var{input_stream_id} within the input 1235file. Both indices start at 0. If specified, 1236@var{sync_file_id}:@var{stream_specifier} sets which input stream 1237is used as a presentation sync reference. 1238 1239The first @code{-map} option on the command line specifies the 1240source for output stream 0, the second @code{-map} option specifies 1241the source for output stream 1, etc. 1242 1243A @code{-} character before the stream identifier creates a "negative" mapping. 1244It disables matching streams from already created mappings. 1245 1246A trailing @code{?} after the stream index will allow the map to be 1247optional: if the map matches no streams the map will be ignored instead 1248of failing. Note the map will still fail if an invalid input file index 1249is used; such as if the map refers to a non-existent input. 1250 1251An alternative @var{[linklabel]} form will map outputs from complex filter 1252graphs (see the @option{-filter_complex} option) to the output file. 1253@var{linklabel} must correspond to a defined output link label in the graph. 1254 1255For example, to map ALL streams from the first input file to output 1256@example 1257ffmpeg -i INPUT -map 0 output 1258@end example 1259 1260For example, if you have two audio streams in the first input file, 1261these streams are identified by "0:0" and "0:1". You can use 1262@code{-map} to select which streams to place in an output file. For 1263example: 1264@example 1265ffmpeg -i INPUT -map 0:1 out.wav 1266@end example 1267will map the input stream in @file{INPUT} identified by "0:1" to 1268the (single) output stream in @file{out.wav}. 1269 1270For example, to select the stream with index 2 from input file 1271@file{a.mov} (specified by the identifier "0:2"), and stream with 1272index 6 from input @file{b.mov} (specified by the identifier "1:6"), 1273and copy them to the output file @file{out.mov}: 1274@example 1275ffmpeg -i a.mov -i b.mov -c copy -map 0:2 -map 1:6 out.mov 1276@end example 1277 1278To select all video and the third audio stream from an input file: 1279@example 1280ffmpeg -i INPUT -map 0:v -map 0:a:2 OUTPUT 1281@end example 1282 1283To map all the streams except the second audio, use negative mappings 1284@example 1285ffmpeg -i INPUT -map 0 -map -0:a:1 OUTPUT 1286@end example 1287 1288To map the video and audio streams from the first input, and using the 1289trailing @code{?}, ignore the audio mapping if no audio streams exist in 1290the first input: 1291@example 1292ffmpeg -i INPUT -map 0:v -map 0:a? OUTPUT 1293@end example 1294 1295To pick the English audio stream: 1296@example 1297ffmpeg -i INPUT -map 0:m:language:eng OUTPUT 1298@end example 1299 1300Note that using this option disables the default mappings for this output file. 1301 1302@item -ignore_unknown 1303Ignore input streams with unknown type instead of failing if copying 1304such streams is attempted. 1305 1306@item -copy_unknown 1307Allow input streams with unknown type to be copied instead of failing if copying 1308such streams is attempted. 1309 1310@item -map_channel [@var{input_file_id}.@var{stream_specifier}.@var{channel_id}|-1][?][:@var{output_file_id}.@var{stream_specifier}] 1311Map an audio channel from a given input to an output. If 1312@var{output_file_id}.@var{stream_specifier} is not set, the audio channel will 1313be mapped on all the audio streams. 1314 1315Using "-1" instead of 1316@var{input_file_id}.@var{stream_specifier}.@var{channel_id} will map a muted 1317channel. 1318 1319A trailing @code{?} will allow the map_channel to be 1320optional: if the map_channel matches no channel the map_channel will be ignored instead 1321of failing. 1322 1323For example, assuming @var{INPUT} is a stereo audio file, you can switch the 1324two audio channels with the following command: 1325@example 1326ffmpeg -i INPUT -map_channel 0.0.1 -map_channel 0.0.0 OUTPUT 1327@end example 1328 1329If you want to mute the first channel and keep the second: 1330@example 1331ffmpeg -i INPUT -map_channel -1 -map_channel 0.0.1 OUTPUT 1332@end example 1333 1334The order of the "-map_channel" option specifies the order of the channels in 1335the output stream. The output channel layout is guessed from the number of 1336channels mapped (mono if one "-map_channel", stereo if two, etc.). Using "-ac" 1337in combination of "-map_channel" makes the channel gain levels to be updated if 1338input and output channel layouts don't match (for instance two "-map_channel" 1339options and "-ac 6"). 1340 1341You can also extract each channel of an input to specific outputs; the following 1342command extracts two channels of the @var{INPUT} audio stream (file 0, stream 0) 1343to the respective @var{OUTPUT_CH0} and @var{OUTPUT_CH1} outputs: 1344@example 1345ffmpeg -i INPUT -map_channel 0.0.0 OUTPUT_CH0 -map_channel 0.0.1 OUTPUT_CH1 1346@end example 1347 1348The following example splits the channels of a stereo input into two separate 1349streams, which are put into the same output file: 1350@example 1351ffmpeg -i stereo.wav -map 0:0 -map 0:0 -map_channel 0.0.0:0.0 -map_channel 0.0.1:0.1 -y out.ogg 1352@end example 1353 1354Note that currently each output stream can only contain channels from a single 1355input stream; you can't for example use "-map_channel" to pick multiple input 1356audio channels contained in different streams (from the same or different files) 1357and merge them into a single output stream. It is therefore not currently 1358possible, for example, to turn two separate mono streams into a single stereo 1359stream. However splitting a stereo stream into two single channel mono streams 1360is possible. 1361 1362If you need this feature, a possible workaround is to use the @emph{amerge} 1363filter. For example, if you need to merge a media (here @file{input.mkv}) with 2 1364mono audio streams into one single stereo channel audio stream (and keep the 1365video stream), you can use the following command: 1366@example 1367ffmpeg -i input.mkv -filter_complex "[0:1] [0:2] amerge" -c:a pcm_s16le -c:v copy output.mkv 1368@end example 1369 1370To map the first two audio channels from the first input, and using the 1371trailing @code{?}, ignore the audio channel mapping if the first input is 1372mono instead of stereo: 1373@example 1374ffmpeg -i INPUT -map_channel 0.0.0 -map_channel 0.0.1? OUTPUT 1375@end example 1376 1377@item -map_metadata[:@var{metadata_spec_out}] @var{infile}[:@var{metadata_spec_in}] (@emph{output,per-metadata}) 1378Set metadata information of the next output file from @var{infile}. Note that 1379those are file indices (zero-based), not filenames. 1380Optional @var{metadata_spec_in/out} parameters specify, which metadata to copy. 1381A metadata specifier can have the following forms: 1382@table @option 1383@item @var{g} 1384global metadata, i.e. metadata that applies to the whole file 1385 1386@item @var{s}[:@var{stream_spec}] 1387per-stream metadata. @var{stream_spec} is a stream specifier as described 1388in the @ref{Stream specifiers} chapter. In an input metadata specifier, the first 1389matching stream is copied from. In an output metadata specifier, all matching 1390streams are copied to. 1391 1392@item @var{c}:@var{chapter_index} 1393per-chapter metadata. @var{chapter_index} is the zero-based chapter index. 1394 1395@item @var{p}:@var{program_index} 1396per-program metadata. @var{program_index} is the zero-based program index. 1397@end table 1398If metadata specifier is omitted, it defaults to global. 1399 1400By default, global metadata is copied from the first input file, 1401per-stream and per-chapter metadata is copied along with streams/chapters. These 1402default mappings are disabled by creating any mapping of the relevant type. A negative 1403file index can be used to create a dummy mapping that just disables automatic copying. 1404 1405For example to copy metadata from the first stream of the input file to global metadata 1406of the output file: 1407@example 1408ffmpeg -i in.ogg -map_metadata 0:s:0 out.mp3 1409@end example 1410 1411To do the reverse, i.e. copy global metadata to all audio streams: 1412@example 1413ffmpeg -i in.mkv -map_metadata:s:a 0:g out.mkv 1414@end example 1415Note that simple @code{0} would work as well in this example, since global 1416metadata is assumed by default. 1417 1418@item -map_chapters @var{input_file_index} (@emph{output}) 1419Copy chapters from input file with index @var{input_file_index} to the next 1420output file. If no chapter mapping is specified, then chapters are copied from 1421the first input file with at least one chapter. Use a negative file index to 1422disable any chapter copying. 1423 1424@item -benchmark (@emph{global}) 1425Show benchmarking information at the end of an encode. 1426Shows real, system and user time used and maximum memory consumption. 1427Maximum memory consumption is not supported on all systems, 1428it will usually display as 0 if not supported. 1429@item -benchmark_all (@emph{global}) 1430Show benchmarking information during the encode. 1431Shows real, system and user time used in various steps (audio/video encode/decode). 1432@item -timelimit @var{duration} (@emph{global}) 1433Exit after ffmpeg has been running for @var{duration} seconds in CPU user time. 1434@item -dump (@emph{global}) 1435Dump each input packet to stderr. 1436@item -hex (@emph{global}) 1437When dumping packets, also dump the payload. 1438@item -re (@emph{input}) 1439Read input at native frame rate. Mainly used to simulate a grab device, 1440or live input stream (e.g. when reading from a file). Should not be used 1441with actual grab devices or live input streams (where it can cause packet 1442loss). 1443By default @command{ffmpeg} attempts to read the input(s) as fast as possible. 1444This option will slow down the reading of the input(s) to the native frame rate 1445of the input(s). It is useful for real-time output (e.g. live streaming). 1446@item -vsync @var{parameter} 1447Video sync method. 1448For compatibility reasons old values can be specified as numbers. 1449Newly added values will have to be specified as strings always. 1450 1451@table @option 1452@item 0, passthrough 1453Each frame is passed with its timestamp from the demuxer to the muxer. 1454@item 1, cfr 1455Frames will be duplicated and dropped to achieve exactly the requested 1456constant frame rate. 1457@item 2, vfr 1458Frames are passed through with their timestamp or dropped so as to 1459prevent 2 frames from having the same timestamp. 1460@item drop 1461As passthrough but destroys all timestamps, making the muxer generate 1462fresh timestamps based on frame-rate. 1463@item -1, auto 1464Chooses between 1 and 2 depending on muxer capabilities. This is the 1465default method. 1466@end table 1467 1468Note that the timestamps may be further modified by the muxer, after this. 1469For example, in the case that the format option @option{avoid_negative_ts} 1470is enabled. 1471 1472With -map you can select from which stream the timestamps should be 1473taken. You can leave either video or audio unchanged and sync the 1474remaining stream(s) to the unchanged one. 1475 1476@item -frame_drop_threshold @var{parameter} 1477Frame drop threshold, which specifies how much behind video frames can 1478be before they are dropped. In frame rate units, so 1.0 is one frame. 1479The default is -1.1. One possible usecase is to avoid framedrops in case 1480of noisy timestamps or to increase frame drop precision in case of exact 1481timestamps. 1482 1483@item -async @var{samples_per_second} 1484Audio sync method. "Stretches/squeezes" the audio stream to match the timestamps, 1485the parameter is the maximum samples per second by which the audio is changed. 1486-async 1 is a special case where only the start of the audio stream is corrected 1487without any later correction. 1488 1489Note that the timestamps may be further modified by the muxer, after this. 1490For example, in the case that the format option @option{avoid_negative_ts} 1491is enabled. 1492 1493This option has been deprecated. Use the @code{aresample} audio filter instead. 1494 1495@item -copyts 1496Do not process input timestamps, but keep their values without trying 1497to sanitize them. In particular, do not remove the initial start time 1498offset value. 1499 1500Note that, depending on the @option{vsync} option or on specific muxer 1501processing (e.g. in case the format option @option{avoid_negative_ts} 1502is enabled) the output timestamps may mismatch with the input 1503timestamps even when this option is selected. 1504 1505@item -start_at_zero 1506When used with @option{copyts}, shift input timestamps so they start at zero. 1507 1508This means that using e.g. @code{-ss 50} will make output timestamps start at 150950 seconds, regardless of what timestamp the input file started at. 1510 1511@item -copytb @var{mode} 1512Specify how to set the encoder timebase when stream copying. @var{mode} is an 1513integer numeric value, and can assume one of the following values: 1514 1515@table @option 1516@item 1 1517Use the demuxer timebase. 1518 1519The time base is copied to the output encoder from the corresponding input 1520demuxer. This is sometimes required to avoid non monotonically increasing 1521timestamps when copying video streams with variable frame rate. 1522 1523@item 0 1524Use the decoder timebase. 1525 1526The time base is copied to the output encoder from the corresponding input 1527decoder. 1528 1529@item -1 1530Try to make the choice automatically, in order to generate a sane output. 1531@end table 1532 1533Default value is -1. 1534 1535@item -enc_time_base[:@var{stream_specifier}] @var{timebase} (@emph{output,per-stream}) 1536Set the encoder timebase. @var{timebase} is a floating point number, 1537and can assume one of the following values: 1538 1539@table @option 1540@item 0 1541Assign a default value according to the media type. 1542 1543For video - use 1/framerate, for audio - use 1/samplerate. 1544 1545@item -1 1546Use the input stream timebase when possible. 1547 1548If an input stream is not available, the default timebase will be used. 1549 1550@item >0 1551Use the provided number as the timebase. 1552 1553This field can be provided as a ratio of two integers (e.g. 1:24, 1:48000) 1554or as a floating point number (e.g. 0.04166, 2.0833e-5) 1555@end table 1556 1557Default value is 0. 1558 1559@item -bitexact (@emph{input/output}) 1560Enable bitexact mode for (de)muxer and (de/en)coder 1561@item -shortest (@emph{output}) 1562Finish encoding when the shortest input stream ends. 1563@item -dts_delta_threshold 1564Timestamp discontinuity delta threshold. 1565@item -dts_error_threshold @var{seconds} 1566Timestamp error delta threshold. This threshold use to discard crazy/damaged 1567timestamps and the default is 30 hours which is arbitrarily picked and quite 1568conservative. 1569@item -muxdelay @var{seconds} (@emph{output}) 1570Set the maximum demux-decode delay. 1571@item -muxpreload @var{seconds} (@emph{output}) 1572Set the initial demux-decode delay. 1573@item -streamid @var{output-stream-index}:@var{new-value} (@emph{output}) 1574Assign a new stream-id value to an output stream. This option should be 1575specified prior to the output filename to which it applies. 1576For the situation where multiple output files exist, a streamid 1577may be reassigned to a different value. 1578 1579For example, to set the stream 0 PID to 33 and the stream 1 PID to 36 for 1580an output mpegts file: 1581@example 1582ffmpeg -i inurl -streamid 0:33 -streamid 1:36 out.ts 1583@end example 1584 1585@item -bsf[:@var{stream_specifier}] @var{bitstream_filters} (@emph{output,per-stream}) 1586Set bitstream filters for matching streams. @var{bitstream_filters} is 1587a comma-separated list of bitstream filters. Use the @code{-bsfs} option 1588to get the list of bitstream filters. 1589@example 1590ffmpeg -i h264.mp4 -c:v copy -bsf:v h264_mp4toannexb -an out.h264 1591@end example 1592@example 1593ffmpeg -i file.mov -an -vn -bsf:s mov2textsub -c:s copy -f rawvideo sub.txt 1594@end example 1595 1596@item -tag[:@var{stream_specifier}] @var{codec_tag} (@emph{input/output,per-stream}) 1597Force a tag/fourcc for matching streams. 1598 1599@item -timecode @var{hh}:@var{mm}:@var{ss}SEP@var{ff} 1600Specify Timecode for writing. @var{SEP} is ':' for non drop timecode and ';' 1601(or '.') for drop. 1602@example 1603ffmpeg -i input.mpg -timecode 01:02:03.04 -r 30000/1001 -s ntsc output.mpg 1604@end example 1605 1606@anchor{filter_complex_option} 1607@item -filter_complex @var{filtergraph} (@emph{global}) 1608Define a complex filtergraph, i.e. one with arbitrary number of inputs and/or 1609outputs. For simple graphs -- those with one input and one output of the same 1610type -- see the @option{-filter} options. @var{filtergraph} is a description of 1611the filtergraph, as described in the ``Filtergraph syntax'' section of the 1612ffmpeg-filters manual. 1613 1614Input link labels must refer to input streams using the 1615@code{[file_index:stream_specifier]} syntax (i.e. the same as @option{-map} 1616uses). If @var{stream_specifier} matches multiple streams, the first one will be 1617used. An unlabeled input will be connected to the first unused input stream of 1618the matching type. 1619 1620Output link labels are referred to with @option{-map}. Unlabeled outputs are 1621added to the first output file. 1622 1623Note that with this option it is possible to use only lavfi sources without 1624normal input files. 1625 1626For example, to overlay an image over video 1627@example 1628ffmpeg -i video.mkv -i image.png -filter_complex '[0:v][1:v]overlay[out]' -map 1629'[out]' out.mkv 1630@end example 1631Here @code{[0:v]} refers to the first video stream in the first input file, 1632which is linked to the first (main) input of the overlay filter. Similarly the 1633first video stream in the second input is linked to the second (overlay) input 1634of overlay. 1635 1636Assuming there is only one video stream in each input file, we can omit input 1637labels, so the above is equivalent to 1638@example 1639ffmpeg -i video.mkv -i image.png -filter_complex 'overlay[out]' -map 1640'[out]' out.mkv 1641@end example 1642 1643Furthermore we can omit the output label and the single output from the filter 1644graph will be added to the output file automatically, so we can simply write 1645@example 1646ffmpeg -i video.mkv -i image.png -filter_complex 'overlay' out.mkv 1647@end example 1648 1649To generate 5 seconds of pure red video using lavfi @code{color} source: 1650@example 1651ffmpeg -filter_complex 'color=c=red' -t 5 out.mkv 1652@end example 1653 1654@item -filter_complex_threads @var{nb_threads} (@emph{global}) 1655Defines how many threads are used to process a filter_complex graph. 1656Similar to filter_threads but used for @code{-filter_complex} graphs only. 1657The default is the number of available CPUs. 1658 1659@item -lavfi @var{filtergraph} (@emph{global}) 1660Define a complex filtergraph, i.e. one with arbitrary number of inputs and/or 1661outputs. Equivalent to @option{-filter_complex}. 1662 1663@item -filter_complex_script @var{filename} (@emph{global}) 1664This option is similar to @option{-filter_complex}, the only difference is that 1665its argument is the name of the file from which a complex filtergraph 1666description is to be read. 1667 1668@item -accurate_seek (@emph{input}) 1669This option enables or disables accurate seeking in input files with the 1670@option{-ss} option. It is enabled by default, so seeking is accurate when 1671transcoding. Use @option{-noaccurate_seek} to disable it, which may be useful 1672e.g. when copying some streams and transcoding the others. 1673 1674@item -seek_timestamp (@emph{input}) 1675This option enables or disables seeking by timestamp in input files with the 1676@option{-ss} option. It is disabled by default. If enabled, the argument 1677to the @option{-ss} option is considered an actual timestamp, and is not 1678offset by the start time of the file. This matters only for files which do 1679not start from timestamp 0, such as transport streams. 1680 1681@item -thread_queue_size @var{size} (@emph{input}) 1682This option sets the maximum number of queued packets when reading from the 1683file or device. With low latency / high rate live streams, packets may be 1684discarded if they are not read in a timely manner; raising this value can 1685avoid it. 1686 1687@item -sdp_file @var{file} (@emph{global}) 1688Print sdp information for an output stream to @var{file}. 1689This allows dumping sdp information when at least one output isn't an 1690rtp stream. (Requires at least one of the output formats to be rtp). 1691 1692@item -discard (@emph{input}) 1693Allows discarding specific streams or frames from streams. 1694Any input stream can be fully discarded, using value @code{all} whereas 1695selective discarding of frames from a stream occurs at the demuxer 1696and is not supported by all demuxers. 1697 1698@table @option 1699@item none 1700Discard no frame. 1701 1702@item default 1703Default, which discards no frames. 1704 1705@item noref 1706Discard all non-reference frames. 1707 1708@item bidir 1709Discard all bidirectional frames. 1710 1711@item nokey 1712Discard all frames excepts keyframes. 1713 1714@item all 1715Discard all frames. 1716@end table 1717 1718@item -abort_on @var{flags} (@emph{global}) 1719Stop and abort on various conditions. The following flags are available: 1720 1721@table @option 1722@item empty_output 1723No packets were passed to the muxer, the output is empty. 1724@item empty_output_stream 1725No packets were passed to the muxer in some of the output streams. 1726@end table 1727 1728@item -xerror (@emph{global}) 1729Stop and exit on error 1730 1731@item -max_muxing_queue_size @var{packets} (@emph{output,per-stream}) 1732When transcoding audio and/or video streams, ffmpeg will not begin writing into 1733the output until it has one packet for each such stream. While waiting for that 1734to happen, packets for other streams are buffered. This option sets the size of 1735this buffer, in packets, for the matching output stream. 1736 1737The default value of this option should be high enough for most uses, so only 1738touch this option if you are sure that you need it. 1739 1740@end table 1741 1742As a special exception, you can use a bitmap subtitle stream as input: it 1743will be converted into a video with the same size as the largest video in 1744the file, or 720x576 if no video is present. Note that this is an 1745experimental and temporary solution. It will be removed once libavfilter has 1746proper support for subtitles. 1747 1748For example, to hardcode subtitles on top of a DVB-T recording stored in 1749MPEG-TS format, delaying the subtitles by 1 second: 1750@example 1751ffmpeg -i input.ts -filter_complex \ 1752 '[#0x2ef] setpts=PTS+1/TB [sub] ; [#0x2d0] [sub] overlay' \ 1753 -sn -map '#0x2dc' output.mkv 1754@end example 1755(0x2d0, 0x2dc and 0x2ef are the MPEG-TS PIDs of respectively the video, 1756audio and subtitles streams; 0:0, 0:3 and 0:7 would have worked too) 1757 1758@section Preset files 1759A preset file contains a sequence of @var{option}=@var{value} pairs, 1760one for each line, specifying a sequence of options which would be 1761awkward to specify on the command line. Lines starting with the hash 1762('#') character are ignored and are used to provide comments. Check 1763the @file{presets} directory in the FFmpeg source tree for examples. 1764 1765There are two types of preset files: ffpreset and avpreset files. 1766 1767@subsection ffpreset files 1768ffpreset files are specified with the @code{vpre}, @code{apre}, 1769@code{spre}, and @code{fpre} options. The @code{fpre} option takes the 1770filename of the preset instead of a preset name as input and can be 1771used for any kind of codec. For the @code{vpre}, @code{apre}, and 1772@code{spre} options, the options specified in a preset file are 1773applied to the currently selected codec of the same type as the preset 1774option. 1775 1776The argument passed to the @code{vpre}, @code{apre}, and @code{spre} 1777preset options identifies the preset file to use according to the 1778following rules: 1779 1780First ffmpeg searches for a file named @var{arg}.ffpreset in the 1781directories @file{$FFMPEG_DATADIR} (if set), and @file{$HOME/.ffmpeg}, and in 1782the datadir defined at configuration time (usually @file{PREFIX/share/ffmpeg}) 1783or in a @file{ffpresets} folder along the executable on win32, 1784in that order. For example, if the argument is @code{libvpx-1080p}, it will 1785search for the file @file{libvpx-1080p.ffpreset}. 1786 1787If no such file is found, then ffmpeg will search for a file named 1788@var{codec_name}-@var{arg}.ffpreset in the above-mentioned 1789directories, where @var{codec_name} is the name of the codec to which 1790the preset file options will be applied. For example, if you select 1791the video codec with @code{-vcodec libvpx} and use @code{-vpre 1080p}, 1792then it will search for the file @file{libvpx-1080p.ffpreset}. 1793 1794@subsection avpreset files 1795avpreset files are specified with the @code{pre} option. They work similar to 1796ffpreset files, but they only allow encoder- specific options. Therefore, an 1797@var{option}=@var{value} pair specifying an encoder cannot be used. 1798 1799When the @code{pre} option is specified, ffmpeg will look for files with the 1800suffix .avpreset in the directories @file{$AVCONV_DATADIR} (if set), and 1801@file{$HOME/.avconv}, and in the datadir defined at configuration time (usually 1802@file{PREFIX/share/ffmpeg}), in that order. 1803 1804First ffmpeg searches for a file named @var{codec_name}-@var{arg}.avpreset in 1805the above-mentioned directories, where @var{codec_name} is the name of the codec 1806to which the preset file options will be applied. For example, if you select the 1807video codec with @code{-vcodec libvpx} and use @code{-pre 1080p}, then it will 1808search for the file @file{libvpx-1080p.avpreset}. 1809 1810If no such file is found, then ffmpeg will search for a file named 1811@var{arg}.avpreset in the same directories. 1812 1813@c man end OPTIONS 1814 1815@chapter Examples 1816@c man begin EXAMPLES 1817 1818@section Video and Audio grabbing 1819 1820If you specify the input format and device then ffmpeg can grab video 1821and audio directly. 1822 1823@example 1824ffmpeg -f oss -i /dev/dsp -f video4linux2 -i /dev/video0 /tmp/out.mpg 1825@end example 1826 1827Or with an ALSA audio source (mono input, card id 1) instead of OSS: 1828@example 1829ffmpeg -f alsa -ac 1 -i hw:1 -f video4linux2 -i /dev/video0 /tmp/out.mpg 1830@end example 1831 1832Note that you must activate the right video source and channel before 1833launching ffmpeg with any TV viewer such as 1834@uref{http://linux.bytesex.org/xawtv/, xawtv} by Gerd Knorr. You also 1835have to set the audio recording levels correctly with a 1836standard mixer. 1837 1838@section X11 grabbing 1839 1840Grab the X11 display with ffmpeg via 1841 1842@example 1843ffmpeg -f x11grab -video_size cif -framerate 25 -i :0.0 /tmp/out.mpg 1844@end example 1845 18460.0 is display.screen number of your X11 server, same as 1847the DISPLAY environment variable. 1848 1849@example 1850ffmpeg -f x11grab -video_size cif -framerate 25 -i :0.0+10,20 /tmp/out.mpg 1851@end example 1852 18530.0 is display.screen number of your X11 server, same as the DISPLAY environment 1854variable. 10 is the x-offset and 20 the y-offset for the grabbing. 1855 1856@section Video and Audio file format conversion 1857 1858Any supported file format and protocol can serve as input to ffmpeg: 1859 1860Examples: 1861@itemize 1862@item 1863You can use YUV files as input: 1864 1865@example 1866ffmpeg -i /tmp/test%d.Y /tmp/out.mpg 1867@end example 1868 1869It will use the files: 1870@example 1871/tmp/test0.Y, /tmp/test0.U, /tmp/test0.V, 1872/tmp/test1.Y, /tmp/test1.U, /tmp/test1.V, etc... 1873@end example 1874 1875The Y files use twice the resolution of the U and V files. They are 1876raw files, without header. They can be generated by all decent video 1877decoders. You must specify the size of the image with the @option{-s} option 1878if ffmpeg cannot guess it. 1879 1880@item 1881You can input from a raw YUV420P file: 1882 1883@example 1884ffmpeg -i /tmp/test.yuv /tmp/out.avi 1885@end example 1886 1887test.yuv is a file containing raw YUV planar data. Each frame is composed 1888of the Y plane followed by the U and V planes at half vertical and 1889horizontal resolution. 1890 1891@item 1892You can output to a raw YUV420P file: 1893 1894@example 1895ffmpeg -i mydivx.avi hugefile.yuv 1896@end example 1897 1898@item 1899You can set several input files and output files: 1900 1901@example 1902ffmpeg -i /tmp/a.wav -s 640x480 -i /tmp/a.yuv /tmp/a.mpg 1903@end example 1904 1905Converts the audio file a.wav and the raw YUV video file a.yuv 1906to MPEG file a.mpg. 1907 1908@item 1909You can also do audio and video conversions at the same time: 1910 1911@example 1912ffmpeg -i /tmp/a.wav -ar 22050 /tmp/a.mp2 1913@end example 1914 1915Converts a.wav to MPEG audio at 22050 Hz sample rate. 1916 1917@item 1918You can encode to several formats at the same time and define a 1919mapping from input stream to output streams: 1920 1921@example 1922ffmpeg -i /tmp/a.wav -map 0:a -b:a 64k /tmp/a.mp2 -map 0:a -b:a 128k /tmp/b.mp2 1923@end example 1924 1925Converts a.wav to a.mp2 at 64 kbits and to b.mp2 at 128 kbits. '-map 1926file:index' specifies which input stream is used for each output 1927stream, in the order of the definition of output streams. 1928 1929@item 1930You can transcode decrypted VOBs: 1931 1932@example 1933ffmpeg -i snatch_1.vob -f avi -c:v mpeg4 -b:v 800k -g 300 -bf 2 -c:a libmp3lame -b:a 128k snatch.avi 1934@end example 1935 1936This is a typical DVD ripping example; the input is a VOB file, the 1937output an AVI file with MPEG-4 video and MP3 audio. Note that in this 1938command we use B-frames so the MPEG-4 stream is DivX5 compatible, and 1939GOP size is 300 which means one intra frame every 10 seconds for 29.97fps 1940input video. Furthermore, the audio stream is MP3-encoded so you need 1941to enable LAME support by passing @code{--enable-libmp3lame} to configure. 1942The mapping is particularly useful for DVD transcoding 1943to get the desired audio language. 1944 1945NOTE: To see the supported input formats, use @code{ffmpeg -demuxers}. 1946 1947@item 1948You can extract images from a video, or create a video from many images: 1949 1950For extracting images from a video: 1951@example 1952ffmpeg -i foo.avi -r 1 -s WxH -f image2 foo-%03d.jpeg 1953@end example 1954 1955This will extract one video frame per second from the video and will 1956output them in files named @file{foo-001.jpeg}, @file{foo-002.jpeg}, 1957etc. Images will be rescaled to fit the new WxH values. 1958 1959If you want to extract just a limited number of frames, you can use the 1960above command in combination with the @code{-frames:v} or @code{-t} option, 1961or in combination with -ss to start extracting from a certain point in time. 1962 1963For creating a video from many images: 1964@example 1965ffmpeg -f image2 -framerate 12 -i foo-%03d.jpeg -s WxH foo.avi 1966@end example 1967 1968The syntax @code{foo-%03d.jpeg} specifies to use a decimal number 1969composed of three digits padded with zeroes to express the sequence 1970number. It is the same syntax supported by the C printf function, but 1971only formats accepting a normal integer are suitable. 1972 1973When importing an image sequence, -i also supports expanding 1974shell-like wildcard patterns (globbing) internally, by selecting the 1975image2-specific @code{-pattern_type glob} option. 1976 1977For example, for creating a video from filenames matching the glob pattern 1978@code{foo-*.jpeg}: 1979@example 1980ffmpeg -f image2 -pattern_type glob -framerate 12 -i 'foo-*.jpeg' -s WxH foo.avi 1981@end example 1982 1983@item 1984You can put many streams of the same type in the output: 1985 1986@example 1987ffmpeg -i test1.avi -i test2.avi -map 1:1 -map 1:0 -map 0:1 -map 0:0 -c copy -y test12.nut 1988@end example 1989 1990The resulting output file @file{test12.nut} will contain the first four streams 1991from the input files in reverse order. 1992 1993@item 1994To force CBR video output: 1995@example 1996ffmpeg -i myfile.avi -b 4000k -minrate 4000k -maxrate 4000k -bufsize 1835k out.m2v 1997@end example 1998 1999@item 2000The four options lmin, lmax, mblmin and mblmax use 'lambda' units, 2001but you may use the QP2LAMBDA constant to easily convert from 'q' units: 2002@example 2003ffmpeg -i src.ext -lmax 21*QP2LAMBDA dst.ext 2004@end example 2005 2006@end itemize 2007@c man end EXAMPLES 2008 2009@include config.texi 2010@ifset config-all 2011@ifset config-avutil 2012@include utils.texi 2013@end ifset 2014@ifset config-avcodec 2015@include codecs.texi 2016@include bitstream_filters.texi 2017@end ifset 2018@ifset config-avformat 2019@include formats.texi 2020@include protocols.texi 2021@end ifset 2022@ifset config-avdevice 2023@include devices.texi 2024@end ifset 2025@ifset config-swresample 2026@include resampler.texi 2027@end ifset 2028@ifset config-swscale 2029@include scaler.texi 2030@end ifset 2031@ifset config-avfilter 2032@include filters.texi 2033@end ifset 2034@end ifset 2035 2036@chapter See Also 2037 2038@ifhtml 2039@ifset config-all 2040@url{ffmpeg.html,ffmpeg} 2041@end ifset 2042@ifset config-not-all 2043@url{ffmpeg-all.html,ffmpeg-all}, 2044@end ifset 2045@url{ffplay.html,ffplay}, @url{ffprobe.html,ffprobe}, 2046@url{ffmpeg-utils.html,ffmpeg-utils}, 2047@url{ffmpeg-scaler.html,ffmpeg-scaler}, 2048@url{ffmpeg-resampler.html,ffmpeg-resampler}, 2049@url{ffmpeg-codecs.html,ffmpeg-codecs}, 2050@url{ffmpeg-bitstream-filters.html,ffmpeg-bitstream-filters}, 2051@url{ffmpeg-formats.html,ffmpeg-formats}, 2052@url{ffmpeg-devices.html,ffmpeg-devices}, 2053@url{ffmpeg-protocols.html,ffmpeg-protocols}, 2054@url{ffmpeg-filters.html,ffmpeg-filters} 2055@end ifhtml 2056 2057@ifnothtml 2058@ifset config-all 2059ffmpeg(1), 2060@end ifset 2061@ifset config-not-all 2062ffmpeg-all(1), 2063@end ifset 2064ffplay(1), ffprobe(1), 2065ffmpeg-utils(1), ffmpeg-scaler(1), ffmpeg-resampler(1), 2066ffmpeg-codecs(1), ffmpeg-bitstream-filters(1), ffmpeg-formats(1), 2067ffmpeg-devices(1), ffmpeg-protocols(1), ffmpeg-filters(1) 2068@end ifnothtml 2069 2070@include authors.texi 2071 2072@ignore 2073 2074@setfilename ffmpeg 2075@settitle ffmpeg video converter 2076 2077@end ignore 2078 2079@bye 2080