Recherche avancée

Sur d’autres sites (6412)

• ffmpeg segments only the first part of my audio file

  30 juin 2012, par hammat

  I'm implementing a http live streaming server to send audio file to iOS devices.
  No problem with Apple's tools, mediafilesegmenter, my files are valid and it works fine.

  I'm trying now to segment the same file using ffmpeg. I've downloaded the last stable version which is the 0.10.2 for now.

  Here is how I try to segment my mp3 file :

  ./ffmpeg -re -i input.mp3 -f segment -segment_time 10 -segment_list outputList.m3u8 -acodec libmp3lame -map 0 output%03d.mp3

  It starts the mapping like expected but finish with only one .mp3 file.

  Did I miss something in the process ?
  Thanks in advance.

  edit

  Ok here is my latest command line :

  ffmpeg -i input.mp3 -c:a libmp3lame -b:a 128k -map 0:0 -f segment -segment_time 10 -segment_list outputlist.m3u8 -segment_format mp3 'output%03d.mp3'

  It still gives me only one file but the file is the hole song, not only one part.
  Here is the output of ffmpeg :

  ffmpeg version 0.10.2 Copyright (c) 2000-2012 the FFmpeg developers 
  
  built on Apr 20 2012 07:08:29 with gcc 4.5.2   
  
  configuration: --enable-gpl --enable-version3 --enable-nonfree --enable-postproc --enable-libmp3lame   
  
      libavutil      51. 35.100 / 51. 35.100   
  
      libavcodec     53. 61.100 / 53. 61.100   
  
      libavformat    53. 32.100 /
  
          53. 32.100   
  
      libavdevice    53.  4.100 / 53.  4.100   
  
      libavfilter     2. 61.100 /  2. 61.100   
  
      libswscale      2.  1.100 /  2.  1.100   
  
      libswresample   0.  6.100 /  0.  6.100   
  
      libpostproc    52.  0.100 /  52.  0.100 
  
      [mp3 @ 0x8e4f120] max_analyze_duration 5000000 reached at 5015510
  
  Input #0, mp3, from 'BeachHouse-Myth.mp3':   
  
      Metadata:
  
              title           : Myth
  
              artist          : Beach House
  
              track           : /
  
              album           : Bloom
  
              disc            : /
  
              genre           : Alternative
  
              TSRC            : USSUB1296501   
  
      Duration: 00:04:18.69, start: 0.000000, bitrate: 320 kb/s
  
              Stream #0:0: Audio: mp3, 44100 Hz, stereo, s16, 320 kb/s Output #0, segment, to 'stream%03d.mp3':   Metadata:
  
              title           : Myth
  
              artist          : Beach House
  
              track           : /
  
              album           : Bloom
  
              disc            : /
  
              genre           : Alternative
  
              TSRC            : USSUB1296501
  
              encoder         : Lavf53.32.100
  
              Stream #0:0: Audio: mp3, 44100 Hz, stereo, s16, 128 kb/s
  
  Stream mapping:   
  
      Stream #0:0 -> #0:0 (mp3 -> libmp3lame) 
  
      Press [q] to stop, [?] for help 
  
      Truncating packet of size 1024 to 105ate=   0.0kbits/s    
  
      Truncating packet of size 1024 to 1 
  
      size=       0kB time=00:04:18.71 bitrate=   0.0kbits/s     video:0kB audio:4042kB global headers:0kB muxing overhead -100.000000%

• Video streaming FFMPEG - VLC video starts black

  7 novembre 2020, par xKedar

  I am trying to stream my webcam when an event occurs to another machine.
Giving the fact that FFMPEG needs around 2 seconds since I call it to when it starts streaming I'm running it in background sending everything to a local socket that consumes it until the event happens and then I reverse the data to the other machine

  


  Here I capture the camera

  


  class ffmpegThread (Thread):
    def __init__(self):
        Thread.__init__(self)
     
    def run(self):
        cam, mic = detect_devices()
        command = 'ffmpeg -f dshow -i video='+cam+':audio='+mic+' -profile:v high -pix_fmt yuvj420p -level:v 4.1 -preset ultrafast -tune zerolatency -vcodec libx264 -r 14 -b:v 512k -s 240x160 -acodec aac -ac 2 -ab 32k -ar 44100 -f mpegts -flush_packets 0 udp://127.0.0.1:'+str(config.ffmpeg_port)+'?pkt_size=1316'
        p = Popen(command , stderr=PIPE)
        for line in iter(p.stderr.readline,''):
           if config.end: break
        p.terminate()
        return 0


  


  This is where I receive and forward the video

  


  
class sendVideoThread (Thread):
    def __init__(self, UDPsenderSocket):
        Thread.__init__(self)
        self.UDPsenderSocket = UDPsenderSocket
 
    def run(self):
        sock = socket.socket(socket.AF_INET, socket.SOCK_DGRAM)
        sock.bind(("127.0.0.1", config.ffmpeg_port))
        bufferSize  = 1348
        firstData = sock.recvfrom(bufferSize)
        while True:
            data = sock.recvfrom(bufferSize)
            if config.event == True:
                startTime = time.time()
                self.UDPsenderSocket.sendto(firstData[0], config.address)
                while time.time() - startTime < 14:
                    self.UDPsenderSocket.sendto(data[0], config.address) 
                    data = sock.recvfrom(bufferSize)                    
                config.event= False
            if config.end: sys.exit()



  


  And those are the vlc options on the other machine(not the whole code)

  


      def play(self):
        self.player = vlc.Instance(["--file-caching=0 --network-caching=0"]).media_player_new()
        self.player.set_mrl('udp://@0.0.0.0:'+str(config.vlcPlayer_port))
        self.player.set_hwnd(self.frame.winfo_id())
        self.player.audio_set_mute(False)
        self.player.play()



  


  The video reaches the other machine but it starts black with audio-only for some seconds then it starts working fine. I guess that by sending the video from a random moment(frame) when the event happens it may happen that vlc will miss some information about the video and needs the next frame with information to come in order to show the video, I was looking to something in order to have more frames with that info but I was not able to find it.

  


• swr : split out DSP functions.

  14 juin 2014, par Ronald S. Bultje

  swr : split out DSP functions.
  DSP bits of swri_resample go into their own mini-DSP functions ; DSP
  
  init goes from a per-call branch in multiple_resample to a proper
  
  DSP init routine ; x86 bits go into x86/ ; swri_resample() moves out of
  
  resample_template.c into resample.c because it’s independent of DSP
  
  code or sample type ; multiple_resample() is simplified.
  Signed-off-by : Michael Niedermayer <michaelni@gmx.at>
  

  • [DH] libswresample/Makefile
  • [DH] libswresample/resample.c
  • [DH] libswresample/resample.h
  • [DH] libswresample/resample_dsp.c
  • [DH] libswresample/resample_template.c
  • [DH] libswresample/x86/Makefile
  • [DH] libswresample/x86/resample_x86_dsp.c