Recherche avancée

Sur d’autres sites (12071)

• ffmpeg legitimate decoding errors

  20 juillet 2017, par Gideon Oduro

  My issue is as follows, i’m sending a H.264 encoded video captured with the help of WebRTC over WebSocket. The idea is to perform server side analysis and object tracking.

  navigator.mediaDevices.getUserMedia(constraint).then((stream) => {
  
    isVideoElement(target, stream)
  
    mediaRecorder = recorder(stream, {mimeType: 'video/webm; codecs=H264'})
  
    mediaRecorder.ondataavailable = (blob) => socket.send(blob.data)
  
    mediaRecorder.start('2000');
  
  })

  On the server side, data is being received as ByteBuffer :

  override fun handleBinaryMessage(session: WebSocketSession, msg: BinaryMessage) {
  
      analysis(msg.payload)
  
  }

  Im using the following resources (resource_1, resource_2) to try to convert my ByteBuffer to a OpenCv frame :

  fun startPreview(data: ByteBuffer) {
  
  avcodec_register_all()
  
  val pack = avcodec.AVPacket()
  
  pack.data(BytePointer(data))
  
  avcodec.av_init_packet(pack)
  
  
  
  val videoData = BytePointer(data)
  
  
  
  val codec = avcodec.avcodec_find_decoder(avcodec.AV_CODEC_ID_H264)
  
  val videoCodecContext = avcodec.avcodec_alloc_context3(codec)
  
  
  
  videoCodecContext.width(1280)
  
  videoCodecContext.height(720)
  
  
  
  videoCodecContext.pix_fmt(avutil.AV_PIX_FMT_YUV420P)
  
  videoCodecContext.codec_type(avutil.AVMEDIA_TYPE_VIDEO)
  
  videoCodecContext.extradata(videoData)
  
  videoCodecContext.extradata_size(data.capacity())
  
  videoCodecContext.flags2(videoCodecContext.flags2() or avcodec.CODEC_FLAG2_CHUNKS)
  
  
  
  avcodec.avcodec_open2(videoCodecContext, codec, null as PointerPointer<*>?)
  
  
  
  val decodedFrameLength = avcodec.avcodec_receive_frame(videoCodecContext, avutil.AVFrame())
  
  
  
  println(decodedFrameLength)
  
  
  
  }

  Im then receiving decodedFrameLength of -35 indicating a decoding error, cant figure out how to proceed from here ?

• No audio output using FFmpeg

  26 mars 2022, par John Mergene Arellano

  I am having problem on Live stream output. I am streaming from mobile app to Node JS server to RTMP. Video output of the live stream is working but not the audio. There is no audio output from live stream.

  


  From my client side, I am sending a stream using the Socket.IO library. I captured the video and audio using getUserMedia API.

  


  navigator.mediaDevices.getUserMedia(constraints).then((stream) => {
    window.videoStream = video.srcObject = stream;
    let mediaRecorder = new MediaRecorder(stream, {
        videoBitsPerSecond : 3 * 1024 * 1024
    });
    mediaRecorder.addEventListener('dataavailable', (e) => {
        let data = e.data;
        socket.emit('live', data);
    });
    mediaRecorder.start(1000);
});


  


  Then my server will receive the stream and write it to FFmpeg.

  


  client.on('live', (stream)=>{
   if(ffmpeg)
       ffmpeg.stdin.write(stream);
});


  


  I tried watching the live video in VLC media player. There is a 5 seconds delay and no audio output.

  


  Please see below for FFmpeg options I used :

  


  ffmpeg = this.CHILD_PROCESS.spawn("ffmpeg", [
   '-f',
   'lavfi',
   '-i', 'anullsrc',
   '-i','-',
   '-c:v', 'libx264', '-preset', 'veryfast', '-tune', 'zerolatency',
   '-c:a', 'aac', '-ar', '44100', '-b:a', '64k',
   '-y', //force to overwrite
   '-use_wallclock_as_timestamps', '1', // used for audio sync
   '-async', '1', // used for audio sync
   '-bufsize', '1000',
   '-f',
   'flv',
   `rtmp://127.0.0.1:1935/live/stream` ]);


  


  What is wrong with my setup ? I need to fix the command so that the live stream will output both video and audio.

  


  I tried streaming to youtube RTMP but still no audio. I am expecting to have an output of video and audio from the getUserMedia API.

  


  What is wrong with my setup ? I need to fix the command so that the live stream will output both video and audio.

  


  I tried streaming to youtube RTMP but still no audio. I am expecting to have an output of video and audio from the getUserMedia API.

  


• Error in converting audio file format from ogg to wav [on hold]

  9 juin 2014, par Sumit Bisht

  I am trying to convert an ogg format file that was created using webrtc (html5 usermedia content generated on firefox) and transferred and decoded on the server into a wav file through ffmpeg but am getting this error on cmmand line while trying to convert :

  $ ffmpeg -i 2014-6-5_16-17-54.ogg res1.wav
  
  ffmpeg version 2.0.1 Copyright (c) 2000-2013 the FFmpeg developers
  
    built on May  1 2014 13:12:12 with gcc 4.4.7 (GCC) 20120313 (Red Hat 4.4.7-4)
  
    configuration: --enable-gpl --enable-version3 --enable-shared --enable-nonfree --enable-postproc --enable-libfaac --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libtheora --enable-libvorbis --enable-libvpx --enable-libx264 --enable-libxvid
  
    libavutil      52. 38.100 / 52. 38.100
  
    libavcodec     55. 18.102 / 55. 18.102
  
    libavformat    55. 12.100 / 55. 12.100
  
    libavdevice    55.  3.100 / 55.  3.100
  
    libavfilter     3. 79.101 /  3. 79.101
  
    libswscale      2.  3.100 /  2.  3.100
  
    libswresample   0. 17.102 /  0. 17.102
  
    libpostproc    52.  3.100 / 52.  3.100
  
  Guessed Channel Layout for  Input Stream #0.0 : mono
  
  Input #0, ogg, from '2014-6-5_16-17-54.ogg':
  
    Duration: 00:00:01.84, start: 0.000000, bitrate: 18 kb/s
  
      Stream #0:0: Audio: opus, 48000 Hz, mono
  
      Metadata:
  
        ENCODER         : Mozilla29.0.1
  
  [graph 0 input from stream 0:0 @ 0x18dca20] Invalid sample format (null)
  
  Error opening filters!

  Although, I am able to play the file on server and using the same command, am able to convert .ogg files generated somewhere else. What might be I missing ?

  Edit :
  Here’s the source code that is used to write to the file :

  1) During startup - use the methods of getUserMedia API.

  navigator.getUserMedia({
  
          audio: true,
  
          video: false
  
      }, function(stream) {
  
          audioStream = RecordRTC(stream, {
  
              bufferSize: 16384
  
          });
  
          audioStream.startRecording();

  2) During stopping of the recording - extracting the recorded information.

  function(audioDataURL) {
  
       var audioFile = {};
  
       audioFile = {
  
         contents: audioDataURL
  
       **strong text**};

  On server end, the following code is creating a file from this data :

  dataURL = dataURL.split(',').pop(); // dataURL is the audioDataURL as defined above
  
  fileBuffer = new Buffer(dataURL, 'base64');
  
  fs.writeFileSync(filePath, fileBuffer);