XAudio2学习之混音

XAudio2不仅可以进行采样率转换，还可以进行混音。所谓混音就是将多路音频混合成一路进行输出。混音主要是IXAudio2SubmixVoice进行此项功能。数据由IXAudio2SourceVoice提交后，流向IXAudio2SubmixVoice进行采样率转换（如果需要的话），并且进行音频混合，最后形成一路采样率和IXAudio2MasteringVoice一致的音频流，由IXAudio2MasteringVoice输出到音频输出设备。

混音主要步骤：

1.初始化COM组件。

2.指定特定格式创建Mastering Voice:2通道/48K，32位深。

3.创建一个IXAudio2SubmixVoice对象，格式和mastering voice一致：作用是混合两路音频为一路，并输出到mastering voice。

4.打开多个文件来提供多路音频数据：这里使用两个文件。

5.针对每个文件创建IXAudio2SourceVoice对象，格式为每个文件的格式。

6.针对每个IXAudio2SourceVoice创建IXAudio2SubmixVoice对象，并分别设置输入格式为每个文件对应的格式，分别设置为IXAudio2SourceVoice对象的输出voice，并分别指定IXAudio2SubmixVoice对象的输出voice为第三步创建的用来混音的IXAudio2SubmixVoice对象。主要用来采样率转换，将采样率由文件提供的采样率转为和Mastering Voice一致的采样率。

7.分配内存，分别读取文件数据，并制定到XAUDIO2_BUFFER。

8.分别调用SubmitSourceBuffer将数据提交。

9.分别调用Start来启动Source Voice。

10.等待播放完成。

11.释放对象和内存数据。

主要实现代码如下：WaveFile.h/WaveFile.cpp请到前面文章中寻找地址自行下载。

#pragma once
#include "WaveFile.h"
#include "XAudio2.h"

class VoiceCallback : public IXAudio2VoiceCallback
{
public:
	HANDLE hBufferEndEvent;
	VoiceCallback() : hBufferEndEvent(CreateEvent(NULL, FALSE, FALSE, NULL)){}
	~VoiceCallback(){ CloseHandle(hBufferEndEvent); }

	//Called when the voice has just finished playing a contiguous audio stream.
	void OnStreamEnd() { SetEvent(hBufferEndEvent); }

	//Unused methods are stubs
	void OnVoiceProcessingPassEnd() { }
	void OnVoiceProcessingPassStart(UINT32 SamplesRequired) {    }
	void OnBufferEnd(void * pBufferContext)    { }
	void OnBufferStart(void * pBufferContext) {    }
	void OnLoopEnd(void * pBufferContext) {    }
	void OnVoiceError(void * pBufferContext, HRESULT Error) { }
};


 int main(int argc, char *argv[])
 {
	 HRESULT hr = CoInitializeEx(NULL, COINIT_MULTITHREADED);;//com初始化
 	if (FAILED(hr))
 		return 0;
 
 	IXAudio2 *pEngine = NULL;
 	hr = XAudio2Create(&pEngine);//创建引擎
 	if (FAILED(hr))
 		return 0;
 
	WAVEFORMATEX waveFormatex;//设置主声音的格式
	waveFormatex.nChannels = 2;
	waveFormatex.nSamplesPerSec = 48000;
	waveFormatex.wBitsPerSample = 32;
	waveFormatex.nAvgBytesPerSec = 384000;
	waveFormatex.nBlockAlign = 8;
	waveFormatex.wFormatTag = WAVE_FORMAT_PCM;

 	IXAudio2MasteringVoice *pMasterVoice = NULL;
 	hr = pEngine->CreateMasteringVoice(&pMasterVoice,waveFormatex.nChannels,waveFormatex.nSamplesPerSec);//创建主声音，默认是输出当前扬声器
 	if (FAILED(hr))
 		return 0;

	XAUDIO2_SEND_DESCRIPTOR pSend;
	pSend.pOutputVoice = pMasterVoice;//指定输出为mastering voice
	pSend.Flags = XAUDIO2_SEND_USEFILTER;

	XAUDIO2_VOICE_SENDS pSendList;
	pSendList.pSends = &pSend;
	pSendList.SendCount = 1;

	IXAudio2SubmixVoice *pSubmixVoice = NULL;
	hr = pEngine->CreateSubmixVoice(&pSubmixVoice, waveFormatex.nChannels, waveFormatex.nSamplesPerSec, 0, 0, &pSendList);//指定输出为mastering voice,作用：混音，将两路音频数据混为一路并输出到mastering voice
	if (FAILED(hr))
		return 0;
	
	CWaveFile waveFile1;
	hr = waveFile1.Open(L"C:\\Users\\xdd\\Desktop\\时间都去哪儿了.wav", NULL, WAVEFILE_READ);//加载文件1
	if (FAILED(hr))
		return 0;


	CWaveFile waveFile2;
	hr = waveFile2.Open(L"C:\\Users\\xdd\\Desktop\\1KHz-stero.wav", NULL, WAVEFILE_READ);//加载文件2
	if (FAILED(hr))
		return 0;

	WAVEFORMATEX *waveFormat1 = waveFile1.GetFormat();//获取文件格式1
	WAVEFORMATEX *waveFormat2 = waveFile2.GetFormat();//获取文件格式2

	pSend.pOutputVoice = pSubmixVoice;//指定输出为SubmixVoice

	IXAudio2SubmixVoice *pSubmixVoice1 = NULL;
	hr = pEngine->CreateSubmixVoice(&pSubmixVoice1, waveFormat1->nChannels, waveFormat1->nSamplesPerSec, 0, 0, &pSendList);//指定输出为pSubmixVoice,作用：对文件1的音频数据进行重采样为mastering voice的采样率
	if (FAILED(hr))
		return 0;

	IXAudio2SubmixVoice *pSubmixVoice2 = NULL;
	hr = pEngine->CreateSubmixVoice(&pSubmixVoice2, waveFormat2->nChannels, waveFormat2->nSamplesPerSec, 0, 0, &pSendList);//指定输出为pSubmixVoice,作用：对文件2的音频数据进行重采样为mastering voice的采样率
	if (FAILED(hr))
		return 0;

	pSend.pOutputVoice = pSubmixVoice1;//指定输出为pSubmixVoice1

	VoiceCallback voiceCallBack1;
 	IXAudio2SourceVoice *pSourceVoice1 = NULL;
	hr = pEngine->CreateSourceVoice(&pSourceVoice1, waveFormat1, 0, 1.0f, &voiceCallBack1,&pSendList);//创建源声音，用来提交数据.指定输出为SubmixVoice1
 	if (FAILED(hr))
 		return 0;

	pSend.pOutputVoice = pSubmixVoice2;//指定输出为pSubmixVoice2

	VoiceCallback voiceCallBack2;
	IXAudio2SourceVoice *pSourceVoice2 = NULL;
	hr = pEngine->CreateSourceVoice(&pSourceVoice2, waveFormat2, 0, 1.0f, &voiceCallBack2, &pSendList);//创建源声音，用来提交数据.指定输出为SubmixVoice2
	if (FAILED(hr))
		return 0;
 
 	DWORD size1 = waveFile1.GetSize();//获取文件的大小
 	BYTE *pData1 = new BYTE[size1];//申请内存空间，用于保存数据
 	hr = waveFile1.Read(pData1, size1, &size1);//读取文件内容
 	if (FAILED(hr))
 		return 0;

	DWORD size2 = waveFile2.GetSize();//获取文件的大小
	BYTE *pData2 = new BYTE[size2];//申请内存空间，用于保存数据
	hr = waveFile2.Read(pData2, size2, &size2);//读取文件内容
	if (FAILED(hr))
		return 0;
 
	XAUDIO2_BUFFER buffer1 = {0};//将读取的文件数据，赋值XAUDIO2_BUFFER
 	buffer1.AudioBytes = size1;
 	buffer1.pAudioData = pData1;
	buffer1.Flags = XAUDIO2_END_OF_STREAM;

	XAUDIO2_BUFFER buffer2 = { 0 };//将读取的文件数据，赋值XAUDIO2_BUFFER
	buffer2.AudioBytes = size2;
	buffer2.pAudioData = pData2;
	buffer2.Flags = XAUDIO2_END_OF_STREAM;
 
 	hr = pSourceVoice1->SubmitSourceBuffer(&buffer1);//提交内存数据
 	if (FAILED(hr))
 		return 0;

	hr = pSourceVoice2->SubmitSourceBuffer(&buffer2);//提交内存数据
	if (FAILED(hr))
		return 0;
 
 	hr = pSourceVoice1->Start(0);//启动源声音
 	if (FAILED(hr))
 		return 0;

	hr = pSourceVoice2->Start(0);//启动源声音
	if (FAILED(hr))
		return 0;
 
	XAUDIO2_VOICE_STATE state;
	pSourceVoice1->GetState(&state);//获取状态
	while (state.BuffersQueued)
	{
		WaitForSingleObject(voiceCallBack1.hBufferEndEvent, INFINITE);
		pSourceVoice2->GetState(&state);
		WaitForSingleObject(voiceCallBack2.hBufferEndEvent, INFINITE);
	}
 
 	pMasterVoice->DestroyVoice();//释放资源
	pSubmixVoice->DestroyVoice();//
	pSubmixVoice1->DestroyVoice();//
	pSubmixVoice2->DestroyVoice();//
 	pSourceVoice1->DestroyVoice();//释放资源
	pSourceVoice2->DestroyVoice();//释放资源
 	pEngine->Release();//释放资源
 	CoUninitialize();//释放资源

	delete []pData1;//释放资源
	pData1 = NULL;

	delete[]pData2;//释放资源
	pData2 = NULL;

 	return 0;
 }

交流QQ：1245178753

本文地址：http://blog.csdn.net/u011417605/article/details/51051039

源码下载：http://download.csdn.net/detail/u011417605/9480309