Recherche avancée

Sur d’autres sites (10895)

• fluent-ffmpeg aspect ratio with crop

  9 juillet 2023, par Dürrani

  I am having trouble changing the aspect ratio between 16:9, 1:1 and 9:16 using fluent-ffmpeg
When I try to change from 16:9 to 9:16, I kinda receive a squeezed video but actually I want the extra part should be removed.

  



  I tried this with many combinations :

  



  FFmpeg()
    .input(video)
    .size("608x?")
    .aspect("9:16")
    .output(tempFile)
    .run();

  



  My input 16:9 video 1920x1080


  



  .

  



  My Expected result 9:16 video 608x1080

  



  

  


• ffmeg : audio drifts in merged TS file for separate audio and video TS files

  13 septembre 2021, par Ronnie Marksch

  basic situation

  


  I have two lists of TS files (each list is specified in an m3u8 file).
The TS files in the first list have only video.
The TS files in the second list have only audio.

  


  I combine the TS files from the first list to get the video only file result_video.ts of duration A (see code for this used below).
I combine the TS files from the second list to get the audio only file result_audio.ts of duration B (see code for this used below).
The problem is that A * 76.28% = B whereas I expected both files to have the same length.
I then combine result_video.ts and result_audio.ts and get a result.ts file where the audio drifts from the video heavily and where the audio has gaps at the points where the chunks have been merged.

  


  Question : how to ensure that the audio is not too slow ? or : how to combine the TS files properly.

  


  code for obtaining the two result files

  


  ffmpeg -safe 0 -f concat -i filelist_video.txt -c copy result_video.ts
ffmpeg -safe 0 -f concat -i filelist_audio.txt -c copy result_audio.ts


  


  code for obtaining the final result

  


  ffmpeg.exe -i .\video_merge_0.ts -i .\audio_merge_0.ts -c copy result.ts


  


  test using only one TS file

  


  I copied the first of the audio input TS files to another TS file.
Doing that produced the output below with a muxing overhead of 22.44%, which is suspiciously close to 1-76.28%.

  


  ffmpeg.exe -i audio_chunk1.ts -c copy audio_chunk1_tmp.ts


  


  I get the info :

  


  Output #0, mpegts, to 'audio_chunk1.ts':
  Metadata:
    encoder         : Lavf58.45.100
    Stream #0:0: Audio: aac (LC) ([15][0][0][0] / 0x000F), 48000 Hz, stereo, fltp, 125 kb/s
Stream mapping:
  Stream #0:0 -> #0:0 (copy)
Press [q] to stop, [?] for help
size=     118kB time=00:00:05.99 bitrate= 161.1kbits/s speed=3.48e+03x
video:0kB audio:96kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 22.444508%


  


  using hls

  


  The following now produces files audio_chunk1_tmp.ts, audio_chunk1_tmp0.ts, ... audio_chunk1_tmp2.ts. So for some reason, the single file already gets split. However, ffmpeg shows that the duration of the result is 5.99s for the 5.87s input (which may be ok).

  


  ffmpeg.exe -i audio_chunk1.ts -f hls audio_chunk1_tmp.ts


  


• Révision 18371 : Report de r18366 :

  22 août 2011, par cedric -

  Manque dans l’analyseur de DTD : il ne verifie pas qu’un attribut de type NMTOKEN ou NMTOKENS ne contient que des noms. Utile pour certains attributs des balises d’en-tete dans la DTD XHTML Basic 1.1, mais curieusement pas pour l’attribut "class" qui est clairement un NMTOKENS pourtant. (...)