Recherche avancée

Sur d’autres sites (12198)

• need help configuring ffmpeg to decode raw AAC with android ndk

  24 octobre 2016, par Matt Wolfe

  I’ve got an android app that gets raw AAC bytes from an external device and I want to decode that data but I can’t seem to get the decoder to work, yet ffmpeg seems to work fine for decoding an mp4 file that contains the same audio data (verified with isoviewer). Recently I was able to get this ffmpeg library on android to decode video frames from the same external device but audio won’t seem to work.

  Here is the ffmpeg output for the file with the same data :

  $ ffmpeg -i Video_2000-01-01_0411.mp4
  
  ffmpeg version 2.6.1 Copyright (c) 2000-2015 the FFmpeg developers
  
    built with Apple LLVM version 6.0 (clang-600.0.57) (based on LLVM 3.5svn)
  
    configuration: --prefix=/usr/local/Cellar/ffmpeg/2.6.1 --enable-shared --enable-pthreads --enable-gpl --enable-version3 --enable-hardcoded-tables --enable-avresample --cc=clang --host-cflags= --host-ldflags= --enable-libx264 --enable-libmp3lame --enable-libvo-aacenc --enable-libxvid --enable-vda
  
    libavutil      54. 20.100 / 54. 20.100
  
    libavcodec     56. 26.100 / 56. 26.100
  
    libavformat    56. 25.101 / 56. 25.101
  
    libavdevice    56.  4.100 / 56.  4.100
  
    libavfilter     5. 11.102 /  5. 11.102
  
    libavresample   2.  1.  0 /  2.  1.  0
  
    libswscale      3.  1.101 /  3.  1.101
  
    libswresample   1.  1.100 /  1.  1.100
  
    libpostproc    53.  3.100 / 53.  3.100
  
  Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'AXON_Flex_Video_2000-01-01_0411.mp4':
  
    Metadata:
  
      major_brand     : mp42
  
      minor_version   : 1
  
      compatible_brands: isom3gp43gp5
  
    Duration: 00:00:15.73, start: 0.000000, bitrate: 1134 kb/s
  
      Stream #0:0(eng): Audio: aac (LC) (mp4a / 0x6134706D), 8000 Hz, mono, fltp, 40 kb/s (default)
  
      Metadata:
  
        handler_name    : soun
  
      Stream #0:1(eng): Video: mpeg4 (Simple Profile) (mp4v / 0x7634706D), yuv420p, 640x480 [SAR 1:1 DAR 4:3], 1087 kb/s, 29.32 fps, 26.58 tbr, 90k tbn, 1k tbc (default)
  
      Metadata:
  
        handler_name    : vide

  Here is my ndk code for setting up and decoding the audio :

  jint ffmpeg_init(JNIEnv * env, jobject this) {
  
      audioCodec = avcodec_find_decoder(AV_CODEC_ID_AAC);
  
      if (!audioCodec) {
  
          LOGE("audio codec %d not found", AV_CODEC_ID_AAC);
  
          return -1;
  
      }
  
  
  
      audioContext = avcodec_alloc_context3(audioCodec);
  
      if (!audioContext) {
  
          LOGE("Could not allocate codec context");
  
          return -1;
  
      }
  
  
  
       int openRet = avcodec_open2(audioContext, audioCodec, NULL);
  
          if (openRet < 0) {
  
            LOGE("Could not open codec, error:%d", openRet);
  
            return -1;
  
          }
  
  
  
      audioContext->sample_rate = 8000;
  
      audioContext->channel_layout = AV_CH_LAYOUT_MONO;
  
      audioContext->profile = FF_PROFILE_AAC_LOW;
  
      audioContext->bit_rate = 48 * 1024;
  
      audioContext->sample_fmt = AV_SAMPLE_FMT_FLTP;
  
  
  
    //  unsigned char extradata[] = {0x15, 0x88}; 
  
    //  audioContext->extradata = extradata;
  
    //  audioContext->extradata_size = sizeof(extradata);
  
      audioFrame = av_frame_alloc();
  
      if (!audioFrame) {
  
          LOGE("Could not create audio frame");
  
          return -1;
  
      }
  
  }
  
  
  
  
  
  jint ffmpeg_decodeAudio(JNIEnv *env, jobject this, jbyteArray aacData, jbyteArray output, int offset, int len) {
  
  
  
      LOGI("ffmpeg_decodeAudio()");
  
      char errbuf[128];
  
      AVPacket avpkt = {0};
  
      av_init_packet(&avpkt);
  
      LOGI("av_init_packet()");
  
      int error, got_frame;    
  
      uint8_t* buffer = (uint8_t *) (*env)->GetByteArrayElements(env, aacData,0);
  
      uint8_t* copy = av_malloc(len);  
  
      memcpy(copy, &buffer[offset], len);
  
      av_packet_from_data(&avpkt, copy, len);
  
  
  
  
  
      if ((error = avcodec_decode_audio4(audioContext, audioFrame, &got_frame, &avpkt)) < 0) {
  
          ffmpeg_log_error(error);
  
          av_free_packet(&avpkt);
  
          return error;
  
      }
  
      if (got_frame) {
  
          LOGE("Copying audioFrame->extended_data to output jbytearray, linesize[0]:%d", audioFrame->linesize[0]);
  
          (*env)->SetByteArrayRegion(env, output, 0, audioFrame->linesize[0],  *audioFrame->extended_data);
  
      }
  
  
  
      return 0;
  
  
  
  }

  As you can see I’ve got an init function that opens the decoder and creates the context, these things all work fine, without error. However when I call avcodec_decode_audio4 I get an error :

  FFMPEG error : -1094995529, Invalid data found when processing input

  I’ve tried all sorts of combinations of AVCodecContext properties. I’m not sure which I need to set for the decoder to do it’s job but from reading online I should just need to set the channel layout and the sample_rate (which I’ve tried by themself). I’ve also tried setting the extradata/extradata_size parameters to that which should match the video settings per : http://wiki.multimedia.cx/index.php?title=MPEG-4_Audio
  But no luck.

  Since the device we’re getting packets from sends aac data that have no sound at the beginning (but are valid packets), I’ve tried to just send those since they definitely should decode correctly.

  Here is an example of the initial audio packets that are of silence :

   010c9eb43f21f90fc87e46fff10a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5dffe214b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4bbd1c429696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696978

  Note that the data shown above is just a hex encoding of the data that I’m putting in AVPacket, and it was sent from an external device to the android application. My application doesn’t have direct access to the file though so I need to decode the raw frames/samples as I get them. When I look at the audio track data in isoviewer I can see that the audio track’s first sample is the same data as what I got from the device that contained that file (thus, the external device is just sending me the sample’s raw data). I believe this data can be derived from reading stsz (sample size) box starting at stco (chunk offset) boxes from the mdat box of the file.

  Also, isoviewer shows the esds box as having the following :

  ESDescriptor{esId=0, streamDependenceFlag=0, URLFlag=0, oCRstreamFlag=0, streamPriority=0, URLLength=0, URLString='null', remoteODFlag=0, dependsOnEsId=0, oCREsId=0, decoderConfigDescriptor=DecoderConfigDescriptor{objectTypeIndication=64, streamType=5, upStream=0, bufferSizeDB=513, maxBitRate=32000, avgBitRate=32000, decoderSpecificInfo=null, audioSpecificInfo=AudioSpecificConfig{configBytes=1588, audioObjectType=2 (AAC LC), samplingFrequencyIndex=11 (8000), samplingFrequency=0, channelConfiguration=1, syncExtensionType=0, frameLengthFlag=0, dependsOnCoreCoder=0, coreCoderDelay=0, extensionFlag=0, layerNr=0, numOfSubFrame=0, layer_length=0, aacSectionDataResilienceFlag=false, aacScalefactorDataResilienceFlag=false, aacSpectralDataResilienceFlag=false, extensionFlag3=0}, configDescriptorDeadBytes=, profileLevelIndicationDescriptors=[[]]}, slConfigDescriptor=SLConfigDescriptor{predefined=2}}

  And the binary is this :

  00 00 00 30 65 73 64 73 00 00 00 00 03 80 80 80
  
  1f 00 00 00 04 80 80 80 14 40 15 00 02 01 00 00
  
  7d 00 00 00 7d 00 05 80 80 80 02 15 88 06 01 02

• msvs : Prefer link.exe from the same directory as cl.exe

  26 juillet 2015, par Henrik Gramner

  msvs : Prefer link.exe from the same directory as cl.exe
  /usr/bin/link from coreutils may be located before the MSVS linker in $PATH
  
  which causes linking to fail due to using the wrong binary.
  

  • [DH] configure

• FFMPEG on Ubuntu Trusty [on hold]

  22 août 2015, par Luuk D. Jansen

  Since FFMPEG is removed from the normal distribution, I used http://ppa.launchpad.net/samrog131/ppa/ubuntu/dists/trusty/main/binary-amd64/Packages for my ffmpeg distribution.

  However, that one does not seem to work anymore. Trying to re-add with ’sudo add-apt-repository ppa:samrog131/ppa’ doesn’t work either.

  I found a ffmpeg distribution on https://launchpad.net/ubuntu/+source/ffmpeg, but I cannot find how to add this to be included in apt-get.

  Can anybody point me in the right direction to get ffmpeg to update again ?