Recherche avancée

Sur d’autres sites (13888)

• Concatenate two files while keeping the duration the same as the two files separately

  21 septembre 2020, par John Pollard

  When I calculate the duration of each individual file I want to concatenate I get 10.24 for both. So I figured when I concatenate the two files I would get a duration of file A plus file B or 10.24 + 10.24 giving me a total duration of 20.48 for the combined file. But no matter what command I use to concentrate I cannot get the same duration. Am I doing something wrong ?

  


  ffprobe -i "audioA.mp3" -show_entries format=duration -v quiet -of csv="p=0"
10.24 


  


  ffprobe -i "audioB.mp3" -show_entries format=duration -v quiet -of csv="p=0"
10.24 


  


  Which makes a total of 10.24 + 10.24 = 20.48 seconds

  


  But when I concatenate the files I get a different duration. Here are my different tries.

  


  Try 1

  


  FFMPEG -y -i 'concat:audioA.mp3|audioB.mp3' -map 0:a -codec:a copy -map_metadata -1 output.mp3
ffprobe -i "output.mp3" -show_entries format=duration -v quiet -of csv="p=0"
20.610612


  


  Try 2

  


  FFMPEG -y -i audioA.mp3 -i audioB.mp3 -filter_complex [0:a][1:a]concat=n=2:v=0:a=1 output.mp3
ffprobe -i "output.mp3" -show_entries format=duration -v quiet -of csv="p=0
20.453878


  


  Try 3

  


  FFMPEG -y -i 'concat:audioA.mp3|audioB.mp3' output.mp3
ffprobe -i "output.mp3" -show_entries format=duration -v quiet -of csv="p=0"
20.506122


  


  

  • Is there a command to use to concatenate that will output a file with
the same duration ?

  


  • Is there a way to do that without reencoding ?

  


  • What makes the durations different in the combined files above ?

  


  


• Cross Fade Arbitrary Number of Videos ffmpeg Efficiently

  15 avril 2022, par jippyjoe4

  I have a series of videos named 'cut_xxx.mp4' where xxx represents a number 000 through 999. I want to do a cross fade on an arbitrary number of them to create a compilation, and each fade should last 4 seconds long. Currently, I'm doing this with Python, but I suspect this is not the most efficient way :

  


  import subprocess    
def get_length(filename):
  result = subprocess.run(["ffprobe", "-v", "error", "-show_entries",
                          "format=duration", "-of",
                          "default=noprint_wrappers=1:nokey=1", filename],
    stdout=subprocess.PIPE,
    stderr=subprocess.STDOUT)
  return float(result.stdout)

CROSS_FADE_DURATION = 4

basevideo = 'cut_000.mp4'
for ii in range(total_videos - 1):
  fade_start = math.floor(get_length(basevideo) - CROSS_FADE_DURATION) # new one
  outfile = f'cross_fade_{ii}.mp4'
  append_video = f'cut_{str(ii+1).zfill(3)}.mp4'
  cfcmd = f'ffmpeg -y -i {basevideo} -i {append_video} -filter_complex "xfade=offset={fade_start}:duration={CROSS_FADE_DURATION}" -an {outfile}'
  basevideo = outfile
  subprocess.call(cfcmd)
  print(fade_start)


  


  I specifically remove the audio with -an because I'll add an audio track later. The issue I see here is that I'm compressing the video over and over again with each individual video file I add to the compilation because I'm only adding one video at a time and then re-encoding.

  


  There should be a way to cross fade multiple videos together into a compilation, but I'm not sure what this would look like or how I would get it to work for an arbitrary number of video files of different durations. Any idea on what that monolithic ffmppeg command would look like or how I could automatically generate it given a list of videos and their durations ?

  


• lavf/matroskaenc : use mkv_check_tag_name consistently

  6 septembre 2016, par Rodger Combs

  lavf/matroskaenc : use mkv_check_tag_name consistently
  Previously, we used a different list of checks when deciding whether to
  
  write a set of tags at all than we did when deciding whether to write an
  
  individual tag in the set. This resulted in sometimes writing an empty
  
  tag master and seekhead. Now we use mkv_check_tag_name everywhere, so
  
  if a dictionary is entirely composed of tags we skip, we don’t write a
  
  tag master at all.
  This affected the test file, since "language" was on one list but not
  
  the other, so we were writing an empty tag master there. The test hash
  
  is updated to reflect that change.
  

  • [DH] libavformat/matroskaenc.c
  • [DH] tests/fate/matroska.mak