Question

我有一个Android应用程序从外部设备获取原始AAC字节，我想解码该数据，但我似乎无法让解码器工作，但ffmpeg似乎可以正常解码mp4文件包含相同的音频数据（使用isoviewer验证）。最近我能够在Android上获得这个ffmpeg库来解码来自同一外部设备的视频帧，但音频似乎不起作用。

以下是具有相同数据的文件的ffmpeg输出：

$ 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

以下是我设置和解码音频的ndk代码：

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;

}

正如你所看到的，我有一个init函数可以打开解码器并创建上下文，这些东西都可以正常工作，没有错误。但是当我调用avcodec_decode_audio4时出现错误：

FFMPEG错误：-1094995529，处理输入时发现无效数据

我尝试过AVCodecContext属性的各种组合。我不确定我需要为解码器设置它的工作但是从在线阅读我应该只需要设置通道布局和sample_rate（我自己尝试过）。我还尝试将extradata / extradata_size参数设置为与每个视频设置匹配的参数：http://wiki.multimedia.cx/index.php?title=MPEG-4_Audio 但没有运气。

由于我们从设备获取数据包发送开始时没有声音的aac数据（但是有效数据包），我试图发送它们，因为它们肯定应该正确解码。

以下是静默的初始音频数据包的示例：

 010c9eb43f21f90fc87e46fff10a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5a5dffe214b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4b4bbd1c429696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696969696978

请注意，上面显示的数据只是我放入AVPacket的数据的十六进制编码，它是从外部设备发送到Android应用程序的。我的应用程序没有直接访问该文件，所以我需要解码原始帧/样本，因为我得到它们。当我查看isoviewer中的音轨数据时，我可以看到音频轨道的第一个样本与我从包含该文件的设备获得的数据相同（因此，外部设备只是向我发送样本的原始数据）。我相信这些数据可以从读取stsz（样本大小）框中获得，该框从文件的mdat框中的stco（块偏移）框开始。

此外，isoviewer将esds框显示为具有以下内容：

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}}

二进制是这样的：

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

Answer 1

我发现了上面代码的主要问题。调用avcodec_open2时，解码器会被初始化。因此，我应该在打开之前先设置上下文字段，如下所示：

df1 <- structure(list(id = c(3243L, 3420L, 8428L, 3420L, 9000L),
val = c("A", 
"B", "A", "C", "D")), .Names = c("id", "val"), class = "data.frame",
row.names = c(NA, -5L))

解码器现在正在解码音频而没有错误。

需要帮助配置ffmpeg解码原始AAC与Android ndk

1 个答案: