Opus解码音频API

创建音频解码器

本节主要讲右手边的音频播放线程的。

代码分析

AudioDecoderOpusImpl::AudioDecoderOpusImpl

AudioDecoderOpusImpl::AudioDecoderOpusImpl(size_t num_channels,
                                           int sample_rate_hz)
    : channels_{num_channels}, sample_rate_hz_{sample_rate_hz} {
  RTC_DCHECK(num_channels == 1 || num_channels == 2);
  RTC_DCHECK(sample_rate_hz == 16000 || sample_rate_hz == 48000);
  const int error =
      WebRtcOpus_DecoderCreate(&dec_state_, channels_, sample_rate_hz_);
  RTC_DCHECK(error == 0);
  WebRtcOpus_DecoderInit(dec_state_);
}

NetEqImpl::DecodeLoop

H:\webrtc-20210315\webrtc-20210315\webrtc\webrtc-checkout\src\modules\audio_coding\neteq\neteq_impl.cc

int NetEqImpl::DecodeLoop(PacketList* packet_list,
                          const Operation& operation,
                          AudioDecoder* decoder,
                          int* decoded_length,
                          AudioDecoder::SpeechType* speech_type) {
  RTC_DCHECK(last_decoded_timestamps_.empty());
  RTC_DCHECK(last_decoded_packet_infos_.empty());
  // Do decoding.
  while (!packet_list->empty() && !decoder_database_->IsComfortNoise(
                                      packet_list->front().payload_type)) {
    assert(decoder);  // At this point, we must have a decoder object.
    // The number of channels in the |sync_buffer_| should be the same as the
    // number decoder channels.
    assert(sync_buffer_->Channels() == decoder->Channels());
    assert(decoded_buffer_length_ >= kMaxFrameSize * decoder->Channels());
    assert(operation == Operation::kNormal ||
           operation == Operation::kAccelerate ||
           operation == Operation::kFastAccelerate ||
           operation == Operation::kMerge ||
           operation == Operation::kPreemptiveExpand);
    // 对传入的音频进行解码操作
    auto opt_result = packet_list->front().frame->Decode(
        rtc::ArrayView<int16_t>(&decoded_buffer_[*decoded_length],
                                decoded_buffer_length_ - *decoded_length));
    last_decoded_timestamps_.push_back(packet_list->front().timestamp);
    last_decoded_packet_infos_.push_back(
        std::move(packet_list->front().packet_info));
    packet_list->pop_front();
    if (opt_result) {
      const auto& result = *opt_result;
      *speech_type = result.speech_type;
      if (result.num_decoded_samples > 0) {
        *decoded_length += rtc::dchecked_cast<int>(result.num_decoded_samples);
        // Update |decoder_frame_length_| with number of samples per channel.
        decoder_frame_length_ =
            result.num_decoded_samples / decoder->Channels();
      }
    } else {
      // Error.
      // TODO(ossu): What to put here?
      RTC_LOG(LS_WARNING) << "Decode error";
      *decoded_length = -1;
      last_decoded_packet_infos_.clear();
      packet_list->clear();
      break;
    }
    if (*decoded_length > rtc::dchecked_cast<int>(decoded_buffer_length_)) {
      // Guard against overflow.
      RTC_LOG(LS_WARNING) << "Decoded too much.";
      packet_list->clear();
      return kDecodedTooMuch;
    }
  }  // End of decode loop.
  // If the list is not empty at this point, either a decoding error terminated
  // the while-loop, or list must hold exactly one CNG packet.
  assert(packet_list->empty() || *decoded_length < 0 ||
         (packet_list->size() == 1 && decoder_database_->IsComfortNoise(
                                          packet_list->front().payload_type)));
  return 0;
}

-》

webrtc::OpusFrame::Decode

absl::optional<DecodeResult> Decode(
      rtc::ArrayView<int16_t> decoded) const override {
    AudioDecoder::SpeechType speech_type = AudioDecoder::kSpeech;
    int ret;
    if (is_primary_payload_) {
      ret = decoder_->Decode(
          payload_.data(), payload_.size(), decoder_->SampleRateHz(),
          decoded.size() * sizeof(int16_t), decoded.data(), &speech_type);
    } else {
      ret = decoder_->DecodeRedundant(
          payload_.data(), payload_.size(), decoder_->SampleRateHz(),
          decoded.size() * sizeof(int16_t), decoded.data(), &speech_type);
    }
    if (ret < 0)
      return absl::nullopt;
    return DecodeResult{static_cast<size_t>(ret), speech_type};
  }

—》

AudioDecoder::Decode

int AudioDecoder::Decode(const uint8_t* encoded,
                         size_t encoded_len,
                         int sample_rate_hz,
                         size_t max_decoded_bytes,
                         int16_t* decoded,
                         SpeechType* speech_type) {
  TRACE_EVENT0("webrtc", "AudioDecoder::Decode");
  rtc::MsanCheckInitialized(rtc::MakeArrayView(encoded, encoded_len));
  int duration = PacketDuration(encoded, encoded_len);
  if (duration >= 0 &&
      duration * Channels() * sizeof(int16_t) > max_decoded_bytes) {
    return -1;
  }
  return DecodeInternal(encoded, encoded_len, sample_rate_hz, decoded,
                        speech_type);
}

AudioDecoderOpusImpl::DecodeInternal

int AudioDecoderOpusImpl::DecodeInternal(const uint8_t* encoded,
                                         size_t encoded_len,
                                         int sample_rate_hz,
                                         int16_t* decoded,
                                         SpeechType* speech_type) {
  RTC_DCHECK_EQ(sample_rate_hz, sample_rate_hz_);
  int16_t temp_type = 1;  // Default is speech.
  int ret =
      WebRtcOpus_Decode(dec_state_, encoded, encoded_len, decoded, &temp_type);
  if (ret > 0)
    ret *= static_cast<int>(channels_);  // Return total number of samples.
    // temp_type==0，返回 AudioDecoder::SpeechType::kSpeech
  *speech_type = ConvertSpeechType(temp_type);
  return ret;
}