Recherche avancée

Sur d’autres sites (7793)

• Is there an elegant way to split a file by chapter using ffmpeg ?

  18 janvier, par Kattern

  In this page, Albert Armea share a code to split videos by chapter using ffmpeg. The code is straight forward, but not quite good-looking.

  


  ffmpeg -i "$SOURCE.$EXT" 2>&1 |
grep Chapter |
sed -E "s/ *Chapter #([0-9]+\.[0-9]+): start ([0-9]+\.[0-9]+), end ([0-9]+\.[0-9]+)/-i \"$SOURCE.$EXT\" -vcodec copy -acodec copy -ss \2 -to \3 \"$SOURCE-\1.$EXT\"/" |
xargs -n 11 ffmpeg


  


  Is there an elegant way to do this job ?

  


• FFMPEG split audio files accurately

  5 mai 2015, par Jakob Hougaard Andersen

  I am trying to use ffmpeg to split uncompressed audio files. I would like to split them very precisely at certain points.
  My experiments so far have led me to this procedure :

  ffmpeg -ss 1.126 -i someInputFile.wav -acodec copy -t 0.634 someOutputFile.wav

  So I am seeking (-ss) to a certain point in the input file and then I am outputting to a file with a defined length (-t).

  The -ss parameter seems to locate the starting point very accurately, but the length of the file doesn’t seem to match my defined length exactly.
  It seems that the file size jumps in steps of 4096 bytes (and the length with it) so that I can not define a file length in between two steps.

  I know that 4096 bytes is not a lot, but for a mono wave file (44.1kHz, 16 bit) it equals a step size of approximately 45 ms. I would really like to be able to define the length as precisely as the starting point.

  So my question is : is it possible to avoid this 4096 byte quatization on the output file length ?

  I have tried to use the ’chomp’ bitstream filter, and it seems to make the length exactly what it should be, but it also causes the output audio file to have strange regions with pure noise...

  Best regards, Jakob

• Filter complex with split source and multiple overlays : Can my code be simplified ?

  29 juin 2024, par Patrick Hennessey

  I've created a complex split filter that splits a single 1372 x 1372 input source into multiple uniquely shaped and cropped slices (s1, s2, etc), and overlays them on a padded background plate into a single output. It also applies a 20fps target framerate on the last overlay step.

  


  It works exactly how I want, but I'm wondering if this code is inefficient or redundant in any way :

  


  ffmpeg -i test.mp4 -filter_complex "[0:v]split=5[s1][s2][s3][s4][s5];
[s1]scale=377:377,crop=360:360:2:2,pad=1920:1080:1560:720[bg];
[s2]crop=1372:1068:0:0[s2];[bg][s2]overlay=0:0[bg];
[s3]crop=460:308:0:1064[s3];[bg][s3]overlay=1372:0[bg];
[s4]crop=460:308:456:1064[s4];[bg][s4]overlay=1372:308[bg];
[s5]crop=460:308:912:1064[s5];[bg][s5]overlay=1372:616,fps=20" output.mp4


  


  Is there a more elegant way to achieve the same result ?