前言

代码分析

创建AudioDeviceBuffer的流程

void WebRtcVoiceEngine::Init() {
    ***
// 4、创建ADM
#if defined(WEBRTC_INCLUDE_INTERNAL_AUDIO_DEVICE)
  // No ADM supplied? Create a default one.
  if (!adm_) {
    adm_ = webrtc::AudioDeviceModule::Create(
        webrtc::AudioDeviceModule::kPlatformDefaultAudio, task_queue_factory_);
  }
#endif  // WEBRTC_INCLUDE_INTERNAL_AUDIO_DEVICE
  RTC_CHECK(adm()); 
  ****
}

—》

rtc::scoped_refptr<AudioDeviceModule> AudioDeviceModule::Create(
    AudioLayer audio_layer,
    TaskQueueFactory* task_queue_factory) {
  RTC_DLOG(INFO) << __FUNCTION__;
  return AudioDeviceModule::CreateForTest(audio_layer, task_queue_factory);
}

—》

rtc::scoped_refptr<AudioDeviceModuleForTest> AudioDeviceModule::CreateForTest(
    AudioLayer audio_layer,
    TaskQueueFactory* task_queue_factory) {
****
  // Create the generic reference counted (platform independent) implementation.
  rtc::scoped_refptr<AudioDeviceModuleImpl> audioDevice(
      new rtc::RefCountedObject<AudioDeviceModuleImpl>(audio_layer,
                                                       task_queue_factory));
**
}

—》

AudioDeviceModuleImpl::AudioDeviceModuleImpl(
    AudioLayer audio_layer,
    TaskQueueFactory* task_queue_factory)
    : audio_layer_(audio_layer), audio_device_buffer_(task_queue_factory) {
  RTC_DLOG(INFO) << __FUNCTION__;
}
AudioDeviceBuffer audio_device_buffer_;

—》

AudioDeviceBuffer::AudioDeviceBuffer(TaskQueueFactory* task_queue_factory)
    : task_queue_(task_queue_factory->CreateTaskQueue(
          kTimerQueueName,
          TaskQueueFactory::Priority::NORMAL)),
      audio_transport_cb_(nullptr),
      rec_sample_rate_(0),
      play_sample_rate_(0),
      rec_channels_(0),
      play_channels_(0),
      playing_(false),
      recording_(false),
      typing_status_(false),
      play_delay_ms_(0),
      rec_delay_ms_(0),
      num_stat_reports_(0),
      last_timer_task_time_(0),
      rec_stat_count_(0),
      play_stat_count_(0),
      play_start_time_(0),
      only_silence_recorded_(true),
      log_stats_(false) {
  RTC_LOG(INFO) << "AudioDeviceBuffer::ctor";
#ifdef AUDIO_DEVICE_PLAYS_SINUS_TONE
  phase_ = 0.0;
  RTC_LOG(WARNING) << "AUDIO_DEVICE_PLAYS_SINUS_TONE is defined!";
#endif
}

注册回调RegisterAudioCallback

void WebRtcVoiceEngine::Init() {
 ***
  // 7、 Connect the ADM to our audio path. 
  // 注册回调，采集到音频数据后，由audio_transport()传输
  adm()->RegisterAudioCallback(audio_state()->audio_transport());
  ***
}

—》

int32_t AudioDeviceModuleImpl::RegisterAudioCallback(
    AudioTransport* audioCallback) {
  RTC_LOG(INFO) << __FUNCTION__;
  return audio_device_buffer_.RegisterAudioCallback(audioCallback);
}

—》

int32_t AudioDeviceBuffer::RegisterAudioCallback(
    AudioTransport* audio_callback) {
  RTC_DCHECK_RUN_ON(&main_thread_checker_);
  RTC_DLOG(INFO) << __FUNCTION__;
  if (playing_ || recording_) {
    RTC_LOG(LS_ERROR) << "Failed to set audio transport since media was active";
    return -1;
  }
  audio_transport_cb_ = audio_callback;
  return 0;
}

传递数据

就是采集到音频数据后，调用上一步的回调。


AudioDeviceWindowsCore::DoCaptureThreadPollDMO

DWORD AudioDeviceWindowsCore::DoCaptureThreadPollDMO() {
****
      ULONG bytesProduced = 0;
      BYTE* data;
      // Get a pointer to the data buffer. This should be valid until
      // the next call to ProcessOutput.  
      // 获取buffer数据和数据大小
      //  bytesProduced=320，因为设置输出采样率16000，每个采样2个字节，每次采集10ms数据
      hr = _mediaBuffer->GetBufferAndLength(&data, &bytesProduced);
      if (FAILED(hr)) {
        _TraceCOMError(hr);
        keepRecording = false;
        assert(false);
        break;
      }
      if (bytesProduced > 0) {
        //断点值 160 = 320/2
        const int kSamplesProduced = bytesProduced / _recAudioFrameSize;
        // TODO(andrew): verify that this is always satisfied. It might
        // be that ProcessOutput will try to return more than 10 ms if
        // we fail to call it frequently enough.
        assert(kSamplesProduced == static_cast<int>(_recBlockSize));
        assert(sizeof(BYTE) == sizeof(int8_t));
        _ptrAudioBuffer->SetRecordedBuffer(reinterpret_cast<int8_t*>(data),
                                           kSamplesProduced);
        _ptrAudioBuffer->SetVQEData(0, 0);
        _UnLock();  // Release lock while making the callback.
         // 将数据输送到上层逻辑去处理
        _ptrAudioBuffer->DeliverRecordedData();
        _Lock();
      }
   ****
  }

—》

int32_t AudioDeviceBuffer::DeliverRecordedData() {
  if (!audio_transport_cb_) {
    RTC_LOG(LS_WARNING) << "Invalid audio transport";
    return 0;
  }
  const size_t frames = rec_buffer_.size() / rec_channels_;
  const size_t bytes_per_frame = rec_channels_ * sizeof(int16_t);
  uint32_t new_mic_level_dummy = 0;
  uint32_t total_delay_ms = play_delay_ms_ + rec_delay_ms_;
  int32_t res = audio_transport_cb_->RecordedDataIsAvailable(
      rec_buffer_.data(), frames, bytes_per_frame, rec_channels_,
      rec_sample_rate_, total_delay_ms, 0, 0, typing_status_,
      new_mic_level_dummy);
  if (res == -1) {
    RTC_LOG(LS_ERROR) << "RecordedDataIsAvailable() failed";
  }
  return 0;
}

—》

// Not used in Chromium. Process captured audio and distribute to all sending
// streams, and try to do this at the lowest possible sample rate.
int32_t AudioTransportImpl::RecordedDataIsAvailable(
    const void* audio_data,
    const size_t number_of_frames,
    const size_t bytes_per_sample,
    const size_t number_of_channels,
    const uint32_t sample_rate,
    const uint32_t audio_delay_milliseconds,
    const int32_t /*clock_drift*/,
    const uint32_t /*volume*/,
    const bool key_pressed,
    uint32_t& /*new_mic_volume*/) {  // NOLINT: to avoid changing APIs
  RTC_DCHECK(audio_data);
  RTC_DCHECK_GE(number_of_channels, 1);
  RTC_DCHECK_LE(number_of_channels, 2);
  RTC_DCHECK_EQ(2 * number_of_channels, bytes_per_sample);
  RTC_DCHECK_GE(sample_rate, AudioProcessing::NativeRate::kSampleRate8kHz);
  // 100 = 1 second / data duration (10 ms).
  RTC_DCHECK_EQ(number_of_frames * 100, sample_rate);
  RTC_DCHECK_LE(bytes_per_sample * number_of_frames * number_of_channels,
                AudioFrame::kMaxDataSizeBytes);
  int send_sample_rate_hz = 0;
  size_t send_num_channels = 0;
  bool swap_stereo_channels = false;
  {
    MutexLock lock(&capture_lock_);
    send_sample_rate_hz = send_sample_rate_hz_;
    send_num_channels = send_num_channels_;
    swap_stereo_channels = swap_stereo_channels_;
  }
  std::unique_ptr<AudioFrame> audio_frame(new AudioFrame());
  InitializeCaptureFrame(sample_rate, send_sample_rate_hz, number_of_channels,
                         send_num_channels, audio_frame.get());
  voe::RemixAndResample(static_cast<const int16_t*>(audio_data),
                        number_of_frames, number_of_channels, sample_rate,
                        &capture_resampler_, audio_frame.get());
  ProcessCaptureFrame(audio_delay_milliseconds, key_pressed,
                      swap_stereo_channels, audio_processing_,
                      audio_frame.get());
  // Typing detection (utilizes the APM/VAD decision). We let the VAD determine
  // if we're using this feature or not.
  // TODO(solenberg): GetConfig() takes a lock. Work around that.
  bool typing_detected = false;
  if (audio_processing_ &&
      audio_processing_->GetConfig().voice_detection.enabled) {
    if (audio_frame->vad_activity_ != AudioFrame::kVadUnknown) {
      bool vad_active = audio_frame->vad_activity_ == AudioFrame::kVadActive;
      typing_detected = typing_detection_.Process(key_pressed, vad_active);
    }
  }
  // Copy frame and push to each sending stream. The copy is required since an
  // encoding task will be posted internally to each stream.
  {
    MutexLock lock(&capture_lock_);
    typing_noise_detected_ = typing_detected;
  }
  RTC_DCHECK_GT(audio_frame->samples_per_channel_, 0);
  if (async_audio_processing_)
    async_audio_processing_->Process(std::move(audio_frame));
  else
    SendProcessedData(std::move(audio_frame));
  return 0;
}

里面调用ProcessCaptureFrame里面是进行音频的3A数据处理。。