本文简要说明最新版WebRtc AudioMixer混音流程。

本程序使用4个16KHz 单声道时长均大于10秒的Wav文件作为混音源，只合成前10秒的音频，输出也是16KHz单声道音频。

输入和输出的采样率都是16000，每10ms音频长度采样点数为160，每个采样点为16bit，两字节大小。

使用的WebRTC代码日期为2019-05-08。

代码如下：

#include "stdafx.h"
#include <string>
#include <iostream>
#include <memory>
#include <thread>
#include <chrono>
#include "webrtc\modules\audio_mixer\audio_mixer_impl.h"
#include "webrtc\api\audio\audio_frame.h"
#include "webrtc\modules\audio_mixer\output_rate_calculator.h"

using namespace std::literals::chrono_literals;

#define WAV_HEADER_SIZE 44
#define SAMPLE_RATE 16000

int16_t s_buf[160] = { 0 };

class AudioSrc : public webrtc::AudioMixer::Source
{
public:
    AudioSrc(int ssrc, int sample, const std::string& file)
        : mSsrc(ssrc)
        , mSample(sample)
        , mFile(nullptr)
    {
        fopen_s(&mFile, file.c_str(), "rb");
        
        if (mFile)
        {
            fseek(mFile, 44, SEEK_SET);
        }
    }

    virtual ~AudioSrc()
    {
        if (mFile)
        {
            fclose(mFile);
            mFile = nullptr;
        }
    }

    virtual AudioFrameInfo GetAudioFrameWithInfo(int sample_rate_hz,
        webrtc::AudioFrame* audio_frame)
    {
        if (mFile)
        {
            fread(s_buf, 160 * 2, 1, mFile);//16KHz, 10ms buf
        }
        
        std::thread::id tid = std::this_thread::get_id();
        std::cout << "thread id ";
        tid._To_text(std::cout);

        //copy s_buf to audio_frame inner buf
        audio_frame->UpdateFrame(0, s_buf, 160, SAMPLE_RATE, webrtc::AudioFrame::SpeechType::kNormalSpeech, webrtc::AudioFrame::VADActivity::kVadUnknown, 1);
        printf(",ssrc %d get audio frame, muted: %d, n %d, s %d\n", mSsrc, int(audio_frame->muted()), audio_frame->num_channels_, audio_frame->sample_rate_hz_);
        
        return AudioFrameInfo::kNormal;
    }

    // A way for a mixer implementation to distinguish participants.
    virtual int Ssrc() const
    {
        return mSsrc;
    }

    // A way for this source to say that GetAudioFrameWithInfo called
    // with this sample rate or higher will not cause quality loss.
    virtual int PreferredSampleRate() const
    {
        return mSample;
    }

private:
    int mSsrc;
    int mSample;
    FILE* mFile;
};


class AudioOutput : public webrtc::OutputRateCalculator
{
    virtual int CalculateOutputRate(
        const std::vector<int>& preferred_sample_rates)
    {
        return SAMPLE_RATE;
    }

};

int MixText()
{
    auto mixptr = webrtc::AudioMixerImpl::Create(std::make_unique<AudioOutput>(), true);
    AudioSrc src1(1, SAMPLE_RATE, "e:\\Media\\Audio\\16k_1.wav");
    AudioSrc src2(2, SAMPLE_RATE, "e:\\Media\\Audio\\16k_2.wav");
    AudioSrc src3(3, SAMPLE_RATE, "e:\\Media\\Audio\\16k_3.wav");
    AudioSrc src4(4, SAMPLE_RATE, "e:\\Media\\Audio\\16k_4.wav");

    mixptr->AddSource(&src1);
    mixptr->AddSource(&src2);
    mixptr->AddSource(&src3);
    mixptr->AddSource(&src4);

    std::thread::id tid = std::this_thread::get_id();
    std::cout << "main thread id: ";
    ti