Recherche avancée

Sur d’autres sites (3940)

• ffmpeg to encode MKV or MP4 with segmented Parts in one file

  28 juillet 2014, par Phil

  Is it Possible to create an MP4 or MKV file where the Video, Meta and Audio Data is segmented and placed alternated in a single File ? (By segmented I don’t mean different Video or Audio Tracks.)

  So that it looks like something like this :
  [meta][video][audio][meta][video][audio]…

  I searched for "segmenting", "clustering", and other buzz words. But the most searches ended up in an HTTP-Livestreaming where One File is segmented in single files.
  That Segmenting Technology is okay (so you can stream it). But I don’t want these segments in single Files, i want them to be in one single MKV or MP4 File.

  Maybe someone has any clue.

• Piwik analytics database : migrating from MySQL to MariaDB

  11 novembre 2015, par Piwik Core Team — Meta

  This short blog post is an announcement regarding the Piwik technology stack.

  Piwik compatible with MySQL and MariaDB

  Since our first public release Piwik uses the open source database server MySQL to store the analytics data.

  Piwik is also compatible with MariaDB. MariaDB is an enhanced, drop-in replacement for MySQL.

  Upgrading to MariaDB

  Many users from our community as well as Piwik PRO have confirmed that using MariaDB for Piwik has several advantages. MariaDB has in some cases significantly improved query performance and reliability of Piwik. Because MariaDB 5.5 is a complete drop-in-replacement for MySQL 5.5, upgrading can be as easy as running apt-get install mariadb-server (or equivalent for your platform). Existing third party techologies such as TokuDB (FAQ) and Galera are fully compatible with MariaDB.

  Learn more about upgrading to MariaDB : Upgrading from MySQL to MariaDB

  In the future, Piwik will stay compatible with both MySQL and MariaDB.

• ffmpeg concatenates mp4s to slowly

  27 août 2022, par Bambi2k21

  I am writing a script that speeds up parts where the speaker does not talk, and normal speed for when he talks. I am done with it, it works fine, but the way I do it is that, I first create all of the sped up mp4's and not sped up(which takes a long time) and then concatenate together. Is there a way to do this without creating new mp4's and spending copious amounts on time.

  


  Code :

  


  for i in range(len(clips)):
    if "a" in clips[i]:
        os.system(r'ffmpeg -i {video_path} -vcodec libx264 -crf 10 -preset ultrafast -ss {first} -to {second} -filter_complex "[0:v]setpts=1*PTS[v];[0:a]atempo=1[a]" -map "[v]" -map "[a]" {cwd}\{third}.mp4'.format(first = clips[i].split("-", 1)[0][1:], second = clips[i].split("-", 1)[1], third = i/10, video_path = video_path, cwd = cwd))
        time.sleep(0.1)
    if "s" in clips[i]:
        os.system(r'ffmpeg -i {video_path} -vcodec libx264 -crf 10 -preset ultrafast -ss {first} -to {second} -filter_complex "[0:v]setpts=1*PTS[v];[0:a]atempo=1[a]" -map "[v]" -map "[a]" {cwd}\{third}.mp4'.format(first = clips[i].split("-", 1)[0][1:], second = clips[i].split("-", 1)[1], third = i/10, video_path = video_path, cwd = cwd))
        time.sleep(0.1)


  


  This is how I create the mp4's.
clips is a list with all of the parts where the speaker talks and doesn't.

  


  os.system(r"(for %i in (*.mp4) do @echo file '%i') > {cwd}\mylist.txt".format(cwd = cwd))
time.sleep(0.1)
os.system("ffmpeg -f concat -i mylist.txt -c copy output.mp4")


  


  This is how I combine the clips. The time sleep is there because without it the clips are not exact and the "time" of each clip is off.