Recherche avancée

Sur d’autres sites (14628)

• find the timestamp of a sound sample of an mp3 with linux or python

  23 juin 2020, par cardamom

  I am slowly working on a project which where it would be very useful if the computer could find where in an mp3 file a certain sample occurs. I would restrict this problem to meaning a fairly exact snippet of the audio, not just for example the chorus in a song on a different recording by the same band where it would become more some kind of machine learning problem. Am thinking if it has no noise added and comes from the same file, it should somehow be possible to locate the time at which it occurs without machine learning, just like grep can find the lines in a textfile where a word occurs.

  


  In case you don't have an mp3 lying around, can set up the problem with some music available on the net which is in the public domain, so nobody complains :

  


  curl https://web.archive.org/web/20041019004300/http://www.navyband.navy.mil/anthems/ANTHEMS/United%20Kingdom.mp3 --output godsavethequeen.mp3


  


  It's a minute long :

  


  exiftool godsavethequeen.mp3 | grep Duration
Duration                        : 0:01:03 (approx)


  


  Now cut out a bit between 30 and 33 seconds (the bit which goes la la la la..) :

  


  ffmpeg -ss 30 -to 33 -i godsavethequeen.mp3 gstq_sample.mp3


  


  both files in the folder :

  


  $ ls -la
-rw-r--r-- 1 cardamom cardamom   48736 Jun 23 00:08 gstq_sample.mp3
-rw-r--r-- 1 cardamom cardamom 1007055 Jun 22 23:57 godsavethequeen.mp3


  


  This is what am after :

  


  $ findsoundsample gstq_sample.mp3 godsavethequeen.mp3
start 30 end 33


  


  Am happy if it is a bash script or a python solution, even using some kind of python library. Sometimes if you use the wrong tool, the solution might work but look horrible, so whichever tool is more suitable. This is a one minute mp3, have not thought yet about performance just about getting it done at all, but would like some scalability, eg find ten seconds somewhere in half an hour.

  


• Unusual results extracting VP9 pkt_size with ffprobe as compared to H.264

  26 novembre 2016, par Jeff S.

  I’m trying to graph the bitrate over time for H.264 and VP9 videos by extracting the frame size with ffprobe, but many of the VP9 videos are showing significantly lower bitrate and total size than both the file size would indicate and that ffprobe reports.

  Can someone point me in the right direction for finding the missing bytes ?

  For example :

  # The extracted values and the ffprobe values are very close for mp4
  
  Video Codec: h264
  
  Video Bitrate: 0.668869
  
  Frame Bitrate: 0.665552571931
  
  Video Size: 6381536.0
  
  Frame Total Size: 6349891
  
  
  
  # The extracted values and the ffprobe values are very different for some vp9 videos
  
  Video Codec: vp9
  
  Video Bitrate: 0.600966
  
  Frame Bitrate: 0.375144984531
  
  Video Size: 5730519.0
  
  Frame Total Size: 3577195

  Below is what I’m using for validation. Note that the sample videos do not contain audio.

  import subprocess
  
  import json
  
  
  
  def export_video_info(video_id):
  
      proc = subprocess.Popen(['ffprobe',
  
          '-v', 'quiet', '-print_format', 
  
          'json', '-show_format',
  
          '-show_streams', '-show_frames',
  
          video_id
  
          ],
  
          stdout=subprocess.PIPE,
  
      )
  
      return proc.communicate()[0]
  
  
  
  # Example video:  https://www.youtube.com/watch?v=g_OdgCrnzYo
  
  # youtube-dl --id -f 135 https://www.youtube.com/watch?v=g_OdgCrnzYo
  
  # youtube-dl --id -f 244 https://www.youtube.com/watch?v=g_OdgCrnzYo
  
  video_list = ['g_OdgCrnzYo.mp4', 'g_OdgCrnzYo.webm']
  
  
  
  '''
  
  Format for every frame: 
  
  
  
      {
  
          "media_type": "video",
  
          "key_frame": 0,
  
          "pkt_pts": 84484,
  
          "pkt_pts_time": "84.484000",
  
          "pkt_dts": 84484,
  
          "pkt_dts_time": "84.484000",
  
          "best_effort_timestamp": 84484,
  
          "best_effort_timestamp_time": "84.484000",
  
          "pkt_duration": 33,
  
          "pkt_duration_time": "0.033000",
  
          "pkt_pos": "7103361",
  
          "pkt_size": "28",
  
          "width": 854,
  
          "height": 480,
  
          "pix_fmt": "yuv420p",
  
          "sample_aspect_ratio": "1:1",
  
          "pict_type": "P",
  
          "coded_picture_number": 0,
  
          "display_picture_number": 0,
  
          "interlaced_frame": 0,
  
          "top_field_first": 0,
  
          "repeat_pict": 0
  
      },
  
  '''
  
  
  
  
  
  # NOTE: videos do not include audio
  
  for video in video_list:
  
      output = json.loads(export_video_info(video))
  
      ff_bitrate = float(output['format']['bit_rate']) / 10**6
  
      ff_duration = float(output['format']['duration'])
  
      ff_codec = output['streams'][0]['codec_name']
  
      ff_size = float(output['format']['size'])
  
      frame_size_sum = 0
  
      for val, items in enumerate(output['frames']):
  
          if output['frames'][val]['media_type'] == 'video': 
  
              frame_size_sum += int(output['frames'][val]['pkt_size'])
  
  
  
      frame_bitrate = frame_size_sum / ff_duration * 8 / 10**6
  
      print('Video Codec: {}\nVideo Bitrate: {}\nFrame Bitrate: {}\nVideo Size: {}\nFrame Total Size: {}\n\n'.format(ff_codec, ff_bitrate, frame_bitrate, ff_size, frame_size_sum))

• upload video stream and audio stream on iOS

  18 juillet 2015, par ronan

  Here’s the needs : People record their video via my App and upload stream to our website, you know, like a live show.

  When they are recording, the network goes bad, then record audio instead of video.
  And when audience watch, first comes the video, in a certain time comes the audio with a static picture.

  How am I supposed to do that ? Thanks.