Recherche avancée

Sur d’autres sites (11066)

• Updated version number.

  16 octobre 2012, par Sebastian Tschan

  m server/php/upload.class.php Updated version number.

• Updated version number.

  3 avril 2012, par Sebastian Tschan

  m server/php/upload.class.php Updated version number.

• Create an audio file and fill it with the RTP package payload using ffmpeg

  24 août 2020, par WithoutExperience

  My task is to extract the payload of an RTP packet and save it as an audio file. In my case, I get the payload type - 8 (PCMA). I process RTP packets and store all the payload in a separate buffer. After that, I have to process this buffer and create an audio file. And that's my problem :(

  


  First, the specification says that in payload 8 (PCMA), audio samples are U8(bitRate / clockRate = 64000 / 8000) in size. But I can't even find a decoder for this sample size. What might be the problem ?

  


  Second, if I create synthetic sound, I can't record it in mka audio format.

  


  As a cheat sheet, I use this link.

  


  Based on this example, I wrote something similar :

  


  
class AudioGenerater
{
public:

    enum AudioFormatType
    {
        AudioFormat_WAV,
        AudioFormat_MP3
    };

public:

    AudioGenerater();
   ~AudioGenerater() = default;

    void generateAudioFileWithOptions(
            QString        fileName,
            QByteArray     pcmData,
            AVCodecID      codecID,
            int            channel,
            int            bitRate,
            int            sampleRate,
            AVSampleFormat format);

private:

    // init Format
    bool initFormat(QString audioFileName);

private:

    AVCodec         *m_AudioCodec        = nullptr;
    AVCodecContext  *m_AudioCodecContext = nullptr;
    AVFormatContext *m_FormatContext     = nullptr;
    AVOutputFormat  *m_OutputFormat      = nullptr;
};



  


  Cpp file :

  


  AudioGenerater::AudioGenerater()
{
    av_register_all();
    avcodec_register_all();
}

bool AudioGenerater::initFormat(QString audioFileName)
{
     // Create an output Format context
    int result = avformat_alloc_output_context2(&m_FormatContext, nullptr, nullptr, audioFileName.toLocal8Bit().data());
    if (result < 0) {
        return false;
    }

    m_OutputFormat = m_FormatContext->oformat;

     // Create an audio stream
    AVStream *audioStream = avformat_new_stream(m_FormatContext, m_AudioCodec);
    if (audioStream == nullptr) {
        avformat_free_context(m_FormatContext);
        return false;
    }

     // Set the parameters in the stream
    audioStream->id = m_FormatContext->nb_streams - 1;
    audioStream->time_base = { 1, 64000 };
    result = avcodec_parameters_from_context(audioStream->codecpar, m_AudioCodecContext);
    if (result < 0) {
        avformat_free_context(m_FormatContext);
        return false;
    }

     // Print FormatContext information
    av_dump_format(m_FormatContext, 0, audioFileName.toLocal8Bit().data(), 1);

     // Open file IO
    if (!(m_OutputFormat->flags & AVFMT_NOFILE)) {
        result = avio_open(&m_FormatContext->pb, audioFileName.toLocal8Bit().data(), AVIO_FLAG_WRITE);
        if (result < 0) {
            avformat_free_context(m_FormatContext);
            return false;
        }
    }

    return true;
}

void AudioGenerater::generateAudioFileWithOptions(
        QString _fileName,
        QByteArray _pcmData,
        AVCodecID _codecID,
        int _channel,
        int _bitRate,
        int _sampleRate,
        AVSampleFormat _format)
{
    AVCodecID codecID = _codecID;
    // Find Codec
    m_AudioCodec = avcodec_find_encoder(codecID);
    if (m_AudioCodec == nullptr) {
        return;
    }

     // Create an encoder context
    m_AudioCodecContext = avcodec_alloc_context3(m_AudioCodec);
    if (m_AudioCodecContext == nullptr){
        return;
    }

        // Setting parameters
    m_AudioCodecContext->bit_rate    = _bitRate;
    m_AudioCodecContext->sample_rate = _sampleRate;
    m_AudioCodecContext->sample_fmt  = _format;
    m_AudioCodecContext->channels    = _channel;


    m_AudioCodecContext->channel_layout = av_get_default_channel_layout(_channel);
    m_AudioCodecContext->flags |= AV_CODEC_FLAG_GLOBAL_HEADER;

     // Turn on the encoder
    int result = avcodec_open2(m_AudioCodecContext, m_AudioCodec, nullptr);
    if (result < 0) {
        avcodec_free_context(&m_AudioCodecContext);
        if (m_FormatContext != nullptr)
            avformat_free_context(m_FormatContext);
        return;
    }

     // Create a package
    if (!initFormat(_fileName)) {
        avcodec_free_context(&m_AudioCodecContext);
        if (m_FormatContext != nullptr)
            avformat_free_context(m_FormatContext);
        return;
    }

     // write to the file header
    result = avformat_write_header(m_FormatContext, nullptr);
    if (result < 0) {
        avcodec_free_context(&m_AudioCodecContext);
        if (m_FormatContext != nullptr)
            avformat_free_context(m_FormatContext);
        return;
    }

     // Create Frame
    AVFrame *frame = av_frame_alloc();
    if (frame == nullptr) {
        avcodec_free_context(&m_AudioCodecContext);
        if (m_FormatContext != nullptr)
            avformat_free_context(m_FormatContext);
        return;
    }

    int nb_samples = 0;
    if (m_AudioCodecContext->codec->capabilities & AV_CODEC_CAP_VARIABLE_FRAME_SIZE) {
        nb_samples = 10000;
    }
    else {
        nb_samples = m_AudioCodecContext->frame_size;
    }

    // Set the parameters of the Frame
    frame->nb_samples = nb_samples;
    frame->format = m_AudioCodecContext->sample_fmt;
    frame->channel_layout = m_AudioCodecContext->channel_layout;

    // Apply for data memory
    result = av_frame_get_buffer(frame, 0);
    if (result < 0)
    {
        av_frame_free(&frame); {
            avcodec_free_context(&m_AudioCodecContext);
            if (m_FormatContext != nullptr)
                avformat_free_context(m_FormatContext);
            return;
        }
    }

    // Set the Frame to be writable
    result = av_frame_make_writable(frame);
    if (result < 0)
    {
        av_frame_free(&frame); {
            avcodec_free_context(&m_AudioCodecContext);
            if (m_FormatContext != nullptr)
                avformat_free_context(m_FormatContext);
            return;
        }
    }

    int perFrameDataSize = frame->linesize[0];
    int count = _pcmData.size() / perFrameDataSize;
    bool needAddOne = false;
    if (_pcmData.size() % perFrameDataSize != 0)
    {
        count++;
        needAddOne = true;
    }

    int frameCount = 0;
    for (int i = 0; i < count; ++i)
    {
         // Create a Packet
        AVPacket *pkt = av_packet_alloc();
        if (pkt == nullptr) {
            avcodec_free_context(&m_AudioCodecContext);
            if (m_FormatContext != nullptr)
                avformat_free_context(m_FormatContext);
            return;
        }
        av_init_packet(pkt);

        if (i == count - 1)
            perFrameDataSize = _pcmData.size() % perFrameDataSize;

        // Synthesize WAV files
       memset(frame->data[0], 0, perFrameDataSize);
       memcpy(frame->data[0], &(_pcmData.data()[perFrameDataSize * i]), perFrameDataSize);


        frame->pts = frameCount++;
         // send Frame
        result = avcodec_send_frame(m_AudioCodecContext, frame);
        if (result < 0)
            continue;

         // Receive the encoded Packet
        result = avcodec_receive_packet(m_AudioCodecContext, pkt);
        if (result < 0)
        {
            av_packet_free(&pkt);
            continue;
        }

         // write to file
        av_packet_rescale_ts(pkt, m_AudioCodecContext->time_base, m_FormatContext->streams[0]->time_base);
        pkt->stream_index = 0;
        result = av_interleaved_write_frame(m_FormatContext, pkt);
        if (result < 0)
            continue;

        av_packet_free(&pkt);
    }

     // write to the end of the file
    av_write_trailer(m_FormatContext);
     // Close file IO
    avio_closep(&m_FormatContext->pb);
    // Release Frame memory
    av_frame_free(&frame);

    avcodec_free_context(&m_AudioCodecContext);
    if (m_FormatContext != nullptr)
        avformat_free_context(m_FormatContext);
}


  


  Main :

  


  int main(int argc, char *argv[])
{
    QCoreApplication a(argc, argv);
    AudioGenerater generator;
    
    av_log_set_level(AV_LOG_DEBUG);
    av_log_set_flags(av_log_get_flags());
    
    QFile file("pcma.raw");
    if (!file.open(QIODevice::ReadOnly)) {
        return EXIT_FAILURE;
    }
    QByteArray pcmaData(file.readAll());
    
    generator.generateAudioFileWithOptions(
                "test.mka",
                pcmaData,
                AV_CODEC_ID_PCM_ALAW,
                1, 64000, 8000,
                AV_SAMPLE_FMT_U8);

    return a.exec();
}



  


  After I run the program, here are the errors that come out :

  


  [pcm_alaw @ 0x19229a0] Specified sample format u8 is invalid or not supported


  


  But if I set s16p instead of U8 format :

  


  generator.generateAudioFileWithOptions(
                "test.mka",
                pcmaData,
                AV_CODEC_ID_PCM_ALAW,
                1, 64000, 8000,
                AV_SAMPLE_FMT_S16P);


  


  I get twice as fast audio track, with a lot of noise. I need a sequential instruction encoding raw audio data into an audio file. Unfortunately, I don't understand what my error is.