Recherche avancée

Sur d’autres sites (4029)

• Pause a FFmpeg encoding in a Python Popen subprocess on Windows

  5 décembre 2020, par CasualDemon

  I am trying to pause an encode of FFmpeg while it is in a non-shell subprocess (This is important to how it plays into a larger program). This can be done by presssing the "Pause / Break" key on the keyboard by itself, and I am trying to send that to Popen.

  


  The command itself must be cross platform compatible, so I cannot wrap it in any way, but I can send signals or run functions that are platform specific as needed.

  


  I looked at how to send a "Ctrl+Break" to a subprocess via pid or handler and it suggested to send a signal, but that raised a "ValueError : Unsupported signal : 21"

  


  from subprocess import Popen, PIPE
import signal


if __name__ == '__main__':
    command = "ffmpeg.exe -y -i example_video.mkv -map 0:v -c:v libx265 -preset slow -crf 18 output.mkv"
    proc = Popen(command, stdin=PIPE, shell=False)

    try:
        proc.send_signal(signal.SIGBREAK)
    finally:
        proc.wait()


  


  Then attempted to use GenerateConsoleCtrlEvent to create a Ctrl+Break event as described here https://docs.microsoft.com/en-us/windows/console/generateconsolectrlevent

  


  from subprocess import Popen, PIPE
import ctypes


if __name__ == '__main__':
    command = "ffmpeg.exe -y -i example_video.mkv -map 0:v -c:v libx265 -preset slow -crf 18 output.mkv"
    proc = Popen(command, stdin=PIPE, shell=False)

    try:
        ctypes.windll.kernel32.GenerateConsoleCtrlEvent(1, proc.pid)
    finally:
        proc.wait()


  


  I have tried psutil pause feature, but it keeps the CPU load really high even when "paused".

  


  Even though it wouldn't work with the program overall, I have at least tried setting creationflags=CREATE_NEW_PROCESS_GROUP which makes the SIGBREAK not error, but also not pause it. For the Ctrl-Break event will entirely stop the encode instead of pausing it.

  


• c# RTSP audio to FFMPEG to SpeechRecognitionEngine

  18 avril 2016, par MrH40XX

  I’m trying to get an audiostream (from any source file/other stream/...) into the microsoft speech recognition engine.

  So far I’ve got :

  ffmpeg.exe -rtsp_transport tcp -i rtsp ://%_return1%/audio -acodec pcm_u16le -f rtp rtp ://localhost:2222

  Then I have inside my code :

  SpeechRecognitionEngine _engine = new SpeechRecognitionEngine(CultureInfo.CurrentCulture);    
  
  this._engine.SetInputToAudioStream(this._rtpClient.AudioStream, new SpeechAudioFormatInfo(16000, AudioBitsPerSample.Sixteen, AudioChannel.Mono));

  Then I have the events registered :

  this._engine.SpeechRecognized += this.SpeechRegocnized;
  
  
  
  this._engine.SpeechDetected += this.EngineOnSpeechDetected;

  I’m not sure about the codec settings... I’ve tried other codecs but doesn’t work.

• Optimizing x264 based remote desktop by dirty regions

  17 novembre 2016, par useprxf

  I was using x264 to achieve remote desktop, but had some problems on handling P_SKIP detection.

  Dirty regions indicate changed areas. For those 16x16 macroblocks which don’t intersect any dirty region, I would like to encode them as P_SKIP macroblocks.

  I tried to add the following code into x264_macroblock_prob_skip_internal function :

  if (! h->isdirty[h->mb.i_mb_x][h->mb.i_mb_y])       // isdirty is a 2-dim array indicating dirty macroblocks
  
      return 1;

  but there is almost no speed-up. I think it may be the information preparation for the macroblock analysis that take influence.

  How to speed up x264 by considering dirty regions ?