kkpcn/CppUtils
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
CppUtils/Media/H264Encoder.cpp
黄 杰 29e3498feb 添加项目文件。
2025-06-05 10:06:43 +08:00

1279 lines
33 KiB
C++
Raw Permalink Blame History

#include "H264Encoder.h"
#include "../Utils/Utils.h"
#pragma comment(lib, "mfreadwrite")
#pragma comment(lib, "mfplat")
#pragma comment(lib, "mfuuid")
#define BREAK_ON_FAIL(value) if(FAILED(value)) break;
#define BREAK_ON_NULL(value, newHr) if(value == NULL) { hr = newHr; break; }
bool H264Encoder::WMFLIB_INITED = false;
static void configWMFLib()
{
if (!H264Encoder::WMFLIB_INITED)
{
CoInitialize(NULL);
CoUninitialize();
CoInitializeEx(NULL, COINIT_APARTMENTTHREADED);
MFStartup(MF_VERSION);
H264Encoder::WMFLIB_INITED = true;
}
}
AudioReader::AudioReader(const std::wstring& filePath) : audioFilename(filePath), m_pSourceReader(NULL)
{
configWMFLib();
HRESULT hr;
IMFMediaType* pPartialMediaType = NULL;
do
{
hr = MFCreateSourceReaderFromURL(audioFilename.c_str(), NULL, &m_pSourceReader);
BREAK_ON_FAIL(hr);
hr = GetSourceInfomation();
BREAK_ON_FAIL(hr);
hr = MFCreateMediaType(&pPartialMediaType);
BREAK_ON_FAIL(hr);
hr = pPartialMediaType->SetGUID(MF_MT_MAJOR_TYPE, MFMediaType_Audio);
BREAK_ON_FAIL(hr);
hr = pPartialMediaType->SetGUID(MF_MT_SUBTYPE, MFAudioFormat_Float);
BREAK_ON_FAIL(hr);
hr = this->SetCurrentMediaType(0, NULL, pPartialMediaType);
pPartialMediaType->Release();
} while (false);
}
AudioReader::AudioReader(const std::vector<BYTE>& fileBuffer) : audioFilename(L""), m_pSourceReader(NULL)
{
configWMFLib();
HRESULT hr;
IMFMediaType* pPartialMediaType = NULL;
do
{
IStream* pStream = SHCreateMemStream(fileBuffer.data(), static_cast<UINT>(fileBuffer.size()));
hr = MFCreateMFByteStreamOnStream(pStream, &pByteStream);
BREAK_ON_FAIL(hr);
pStream->Release();
hr = MFCreateSourceReaderFromByteStream(pByteStream, NULL, &m_pSourceReader);
BREAK_ON_FAIL(hr);
hr = GetSourceInfomation();
BREAK_ON_FAIL(hr);
hr = MFCreateMediaType(&pPartialMediaType);
BREAK_ON_FAIL(hr);
hr = pPartialMediaType->SetGUID(MF_MT_MAJOR_TYPE, MFMediaType_Audio);
BREAK_ON_FAIL(hr);
hr = pPartialMediaType->SetGUID(MF_MT_SUBTYPE, MFAudioFormat_Float);
BREAK_ON_FAIL(hr);
hr = this->SetCurrentMediaType(0, NULL, pPartialMediaType);
pPartialMediaType->Release();
} while (false);
}
AudioReader::AudioReader(IMFMediaSource* mediaSource)
{
configWMFLib();
HRESULT hr;
IMFMediaType* pPartialMediaType = NULL;
do
{
hr = MFCreateSourceReaderFromMediaSource(mediaSource, NULL, &m_pSourceReader);
BREAK_ON_FAIL(hr);
hr = GetSourceInfomation();
BREAK_ON_FAIL(hr);
hr = MFCreateMediaType(&pPartialMediaType);
BREAK_ON_FAIL(hr);
hr = pPartialMediaType->SetGUID(MF_MT_MAJOR_TYPE, MFMediaType_Audio);
BREAK_ON_FAIL(hr);
hr = pPartialMediaType->SetGUID(MF_MT_SUBTYPE, MFAudioFormat_Float);
BREAK_ON_FAIL(hr);
hr = this->SetCurrentMediaType(0, NULL, pPartialMediaType);
pPartialMediaType->Release();
} while (false);
}
AudioReader::~AudioReader() {
if (m_pSourceReader)
m_pSourceReader->Release();
if (pByteStream)
pByteStream->Release();
}
IMFMediaType* AudioReader::GetCurrentMediaType(UINT32 index)
{
IMFMediaType* result = NULL;
HRESULT hr = m_pSourceReader->GetCurrentMediaType(index, &result);
if (FAILED(hr))
return NULL;
return result;
}
HRESULT AudioReader::GetSourceInfomation()
{
EncodingBitRate = NULL;
Duration = NULL;
IMFPresentationDescriptor* pPD = nullptr;
IMFMediaSource* pSource = nullptr;
HRESULT hr = m_pSourceReader->GetServiceForStream(
MF_SOURCE_READER_MEDIASOURCE,
GUID_NULL, IID_IMFMediaSource,
(void**)&pSource
);
if (FAILED(hr))
return hr;
hr = pSource->CreatePresentationDescriptor(&pPD);
if (SUCCEEDED(hr))
{
hr = pPD->GetUINT32(MF_PD_AUDIO_ENCODING_BITRATE, &EncodingBitRate);
hr = pPD->GetUINT64(MF_PD_DURATION, &Duration);
hr = pPD->GetUINT64(MF_PD_TOTAL_FILE_SIZE, &FileSize);
pPD->Release();
IMFMediaType* pMediaType = this->GetCurrentMediaType(0);
hr = pMediaType->GetUINT32(MF_MT_AUDIO_NUM_CHANNELS, &ChannelCount);
hr = pMediaType->GetUINT32(MF_MT_AUDIO_SAMPLES_PER_SECOND, &SamplesPerSecond);
hr = pMediaType->GetUINT32(MF_MT_AUDIO_AVG_BYTES_PER_SECOND, &AverageBytesPerSecond);
hr = pMediaType->GetUINT32(MF_MT_AUDIO_BLOCK_ALIGNMENT, &AudioBlockAlignment);
hr = pMediaType->GetUINT32(MF_MT_AUDIO_BITS_PER_SAMPLE, &AudioBitsPerSample);
hr = pMediaType->GetGUID(MF_MT_SUBTYPE, &SubType);
hr = pMediaType->GetGUID(MF_MT_MAJOR_TYPE, &MediaType);
pMediaType->Release();
pSource->Release();
}
return hr;
}
IMFSample* AudioReader::ReadNextSample() {
LONGLONG timestamp = 0;
DWORD streamIndex;
DWORD sampleflag;
IMFSample* sample;
HRESULT hr = m_pSourceReader->ReadSample(
MF_SOURCE_READER_FIRST_AUDIO_STREAM, 0, &streamIndex, &sampleflag, &timestamp, &sample);
return SUCCEEDED(hr) && sample ? sample : NULL;
}
HRESULT AudioReader::SetPosition(LONGLONG timeStamp)
{
PROPVARIANT var;
PropVariantInit(&var);
var.vt = VT_I8;
var.hVal.QuadPart = timeStamp;
return m_pSourceReader->SetCurrentPosition(GUID_NULL, var);
}
double AudioReader::LenghtOfSeconds() const {
return (double)Duration / 10000000.0;
}
IMFMediaType* AudioReader::GetNativeMediaType(UINT32 index)
{
IMFMediaType* pStreamMediaType = NULL;
HRESULT hr = m_pSourceReader->GetNativeMediaType(
index, 0, &pStreamMediaType);
if (FAILED(hr))
return NULL;
return pStreamMediaType;
}
HRESULT AudioReader::SetCurrentMediaType(DWORD dwStreamIndex, DWORD* pdwReserved, IMFMediaType* pMediaType)
{
return m_pSourceReader->SetCurrentMediaType(dwStreamIndex, pdwReserved, pMediaType);
}
std::vector<MixedFrame> AudioReader::ReadAudioFrames(IMFMediaType* targetMediaType)
{
HRESULT hr = S_OK;
std::vector<MixedFrame> frames;
IMFMediaType* pCurrentMediaType = this->GetCurrentMediaType(0);
AudioResampler* resampler = nullptr;
bool needResample = false;
if (targetMediaType)
{
BOOL isEqual = FALSE;
hr = pCurrentMediaType->Compare(targetMediaType, MF_ATTRIBUTES_MATCH_INTERSECTION, &isEqual);
if (FAILED(hr) || !isEqual)
{
needResample = true;
resampler = new AudioResampler(pCurrentMediaType, targetMediaType);
}
}
pCurrentMediaType->Release();
IMFSample* sample = nullptr;
while ((sample = this->ReadNextSample()) != nullptr)
{
IMFSample* processedSample = sample;
if (needResample)
{
hr = resampler->Resample(sample, &processedSample);
if (FAILED(hr))
{
sample->Release();
break;
}
}
LONGLONG time = 0;
LONGLONG duration = 0;
processedSample->GetSampleTime(&time);
processedSample->GetSampleDuration(&duration);
IMFMediaBuffer* pInputBuffer = nullptr;
hr = processedSample->ConvertToContiguousBuffer(&pInputBuffer);
if (FAILED(hr))
continue;
BYTE* pInputData = nullptr;
DWORD inputDataSize = 0;
hr = pInputBuffer->Lock(&pInputData, NULL, &inputDataSize);
if (FAILED(hr))
continue;
float* pData = (float*)pInputData;
DWORD dataSize = inputDataSize;
size_t sampleCount = dataSize / sizeof(float);
std::vector<float> data(pData, pData + sampleCount);
pInputBuffer->Unlock();
pInputBuffer->Release();
MixedFrame frame;
frame.Data = data;
frame.Time = time;
frame.Duration = duration;
frames.push_back(frame);
if (processedSample != sample)
processedSample->Release();
sample->Release();
}
if (resampler)
delete resampler;
return frames;
}
AudioResampler::AudioResampler(IMFMediaType* pInputType, IMFMediaType* pOutputType)
: m_pResampler(nullptr)
{
HRESULT hr = S_OK;
auto CLSID_CResamplerMediaObject_GUID = Guid::Parse("f447b69e-1884-4a7e-8055-346f74d6edb3").Data;
hr = CoCreateInstance(
CLSID_CResamplerMediaObject_GUID,
nullptr,
CLSCTX_INPROC_SERVER,
IID_PPV_ARGS(&m_pResampler)
);
if (SUCCEEDED(hr))
{
hr = m_pResampler->SetInputType(0, pInputType, 0);
}
if (SUCCEEDED(hr))
{
hr = m_pResampler->SetOutputType(0, pOutputType, 0);
}
}
AudioResampler::~AudioResampler()
{
if (m_pResampler)
{
m_pResampler->Release();
}
}
HRESULT AudioResampler::Resample(IMFSample* pInputSample, IMFSample** ppOutputSample)
{
HRESULT hr = S_OK;
MFT_OUTPUT_DATA_BUFFER outputDataBuffer;
DWORD status = 0;
hr = m_pResampler->ProcessInput(0, pInputSample, 0);
if (SUCCEEDED(hr))
{
ZeroMemory(&outputDataBuffer, sizeof(outputDataBuffer));
outputDataBuffer.pSample = nullptr;
hr = MFCreateSample(&(outputDataBuffer.pSample));
}
if (SUCCEEDED(hr))
{
IMFMediaBuffer* pBuffer = nullptr;
DWORD outputBufferSize = 0;
IMFMediaType* pOutputType = nullptr;
hr = m_pResampler->GetOutputCurrentType(0, &pOutputType);
if (SUCCEEDED(hr))
{
UINT32 sampleRate = 0, channelCount = 0, bitsPerSample = 0;
hr = pOutputType->GetUINT32(MF_MT_AUDIO_SAMPLES_PER_SECOND, &sampleRate);
hr = pOutputType->GetUINT32(MF_MT_AUDIO_NUM_CHANNELS, &channelCount);
hr = pOutputType->GetUINT32(MF_MT_AUDIO_BITS_PER_SAMPLE, &bitsPerSample);
LONGLONG duration = 0;
pInputSample->GetSampleDuration(&duration);
UINT64 numSamples = (UINT64)(sampleRate * duration / 10000000);
outputBufferSize = (DWORD)(numSamples * channelCount * (bitsPerSample / 8));
pOutputType->Release();
}
if (SUCCEEDED(hr))
{
hr = MFCreateMemoryBuffer(outputBufferSize, &pBuffer);
}
if (SUCCEEDED(hr))
{
hr = outputDataBuffer.pSample->AddBuffer(pBuffer);
pBuffer->Release();
}
}
if (SUCCEEDED(hr))
{
hr = m_pResampler->ProcessOutput(0, 1, &outputDataBuffer, &status);
if (hr == MF_E_TRANSFORM_NEED_MORE_INPUT)
{
hr = S_OK;
}
else if (SUCCEEDED(hr))
{
*ppOutputSample = outputDataBuffer.pSample;
(*ppOutputSample)->AddRef();
}
}
if (outputDataBuffer.pSample)
{
outputDataBuffer.pSample->Release();
}
return hr;
}
/*
将两个音频按权重进行混合
mainVolume:主音频音量权重[0-1]
secondVolume:副音频音量权重[0-1]
clipToMainAudio:是否将副音频剪切到匹配主音频的长度
loopMode:如果副音频长度比主音频短则循环副音频
*/
std::vector<MixedFrame> AudioMix::MixAudio(AudioReader* mainAudio, float mainVolume, AudioReader* second, float secondVolume, bool clipToMainAudio, bool loopMode)
{
IMFMediaType* pFirstMediaType = mainAudio->GetCurrentMediaType(0);
std::vector<MixedFrame> framesFirst = mainAudio->ReadAudioFrames(NULL);
std::vector<MixedFrame> framesSecond = second->ReadAudioFrames(pFirstMediaType);
pFirstMediaType->Release();
UINT32 sampleRate = mainAudio->SamplesPerSecond;
UINT32 channels = mainAudio->ChannelCount;
double samplesPerSecond = static_cast<double>(sampleRate * channels);
size_t totalSamplesFirst = 0;
for (const auto& frame : framesFirst)
{
totalSamplesFirst += frame.Data.size();
}
size_t totalSamplesSecond = 0;
for (const auto& frame : framesSecond)
{
totalSamplesSecond += frame.Data.size();
}
size_t totalSamples = max(totalSamplesFirst, totalSamplesSecond);
std::vector<float> mixedData;
mixedData.reserve(totalSamples);
size_t indexFirst = 0, indexSecond = 0;
size_t frameIndexFirst = 0, frameIndexSecond = 0;
size_t sampleIndexFirst = 0, sampleIndexSecond = 0;
float maxVal = 0.0f, minVal = 0.0f;
while (indexFirst < totalSamplesFirst || indexSecond < totalSamplesSecond)
{
float sampleValue = 0.0f;
if (indexFirst < totalSamplesFirst)
{
if (sampleIndexFirst >= framesFirst[frameIndexFirst].Data.size())
{
frameIndexFirst++;
sampleIndexFirst = 0;
}
if (frameIndexFirst < framesFirst.size())
{
sampleValue += framesFirst[frameIndexFirst].Data[sampleIndexFirst++] * mainVolume;
indexFirst++;
}
}
if (loopMode)
{
if (indexFirst >= totalSamplesFirst)
break;
indexSecond = 0;
frameIndexSecond = 0;
sampleIndexSecond = 0;
}
if (indexSecond < totalSamplesSecond)
{
if (sampleIndexSecond >= framesSecond[frameIndexSecond].Data.size())
{
frameIndexSecond++;
sampleIndexSecond = 0;
}
if (frameIndexSecond < framesSecond.size())
{
sampleValue += framesSecond[frameIndexSecond].Data[sampleIndexSecond++] * secondVolume;
indexSecond++;
}
}
if (sampleValue > maxVal)
maxVal = sampleValue;
if (sampleValue < minVal)
minVal = sampleValue;
mixedData.push_back(sampleValue);
if (clipToMainAudio && indexFirst >= totalSamplesFirst)
break;
}
if (maxVal > 0.75f || minVal < -0.75f)
{
float maxMul = max(maxVal, -minVal);
for (auto& fm : mixedData)
{
fm /= maxMul;
}
}
std::vector<MixedFrame> result;
size_t samplesPerFrame = framesFirst[0].Data.size();
size_t totalMixedSamples = mixedData.size();
size_t currentSampleIndex = 0;
size_t currFrameIndex = 0;
while (currentSampleIndex < totalMixedSamples)
{
if (currFrameIndex < framesFirst.size())
samplesPerFrame = framesFirst[currFrameIndex++].Data.size();
MixedFrame frame;
size_t remainingSamples = totalMixedSamples - currentSampleIndex;
size_t currentFrameSamples = min(samplesPerFrame, remainingSamples);
frame.Data.insert(frame.Data.end(),
mixedData.begin() + currentSampleIndex,
mixedData.begin() + currentSampleIndex + currentFrameSamples);
frame.Duration = static_cast<LONGLONG>((currentFrameSamples * 10000000) / samplesPerSecond);
frame.Time = static_cast<LONGLONG>((currentSampleIndex * 10000000) / samplesPerSecond);
result.push_back(frame);
currentSampleIndex += currentFrameSamples;
}
return result;
}
/*
将一个主音频和多个连续的副音频按权重进行混合
mainVolume:主音频音量权重[0-1]
secondVolume:副音频音量权重[0-1]
clipToMainAudio:是否将副音频剪切到匹配主音频的长度
loopMode:如果副音频长度比主音频短则循环副音频
*/
std::vector<MixedFrame> AudioMix::MixAudio(AudioReader* mainAudio, float mainVolume, const std::vector<AudioReader*>& secondList, float secondVolume, bool clipToMainAudio, bool loopMode)
{
IMFMediaType* pFirstMediaType = mainAudio->GetCurrentMediaType(0);
std::vector<MixedFrame> framesFirst = mainAudio->ReadAudioFrames(NULL);
std::vector<MixedFrame> framesSecond;
for (auto sec : secondList)
{
std::vector<MixedFrame> secTmp = sec->ReadAudioFrames(pFirstMediaType);
framesSecond.insert(framesSecond.end(), secTmp.begin(), secTmp.end());
}
pFirstMediaType->Release();
UINT32 sampleRate = mainAudio->SamplesPerSecond;
UINT32 channels = mainAudio->ChannelCount;
double samplesPerSecond = static_cast<double>(sampleRate * channels);
size_t totalSamplesFirst = 0;
for (const auto& frame : framesFirst)
{
totalSamplesFirst += frame.Data.size();
}
size_t totalSamplesSecond = 0;
for (const auto& frame : framesSecond)
{
totalSamplesSecond += frame.Data.size();
}
size_t totalSamples = max(totalSamplesFirst, totalSamplesSecond);
std::vector<float> mixedData;
mixedData.reserve(totalSamples);
size_t indexFirst = 0, indexSecond = 0;
size_t frameIndexFirst = 0, frameIndexSecond = 0;
size_t sampleIndexFirst = 0, sampleIndexSecond = 0;
float maxVal = 0.0f, minVal = 0.0f;
while (indexFirst < totalSamplesFirst || indexSecond < totalSamplesSecond)
{
float sampleValue = 0.0f;
if (indexFirst < totalSamplesFirst)
{
if (sampleIndexFirst >= framesFirst[frameIndexFirst].Data.size())
{
frameIndexFirst++;
sampleIndexFirst = 0;
}
if (frameIndexFirst < framesFirst.size())
{
sampleValue += framesFirst[frameIndexFirst].Data[sampleIndexFirst++] * mainVolume;
indexFirst++;
}
}
if (loopMode && indexSecond >= totalSamplesSecond)
{
indexSecond = 0;
frameIndexSecond = 0;
sampleIndexSecond = 0;
}
if (indexSecond < totalSamplesSecond)
{
if (sampleIndexSecond >= framesSecond[frameIndexSecond].Data.size())
{
frameIndexSecond++;
sampleIndexSecond = 0;
}
if (frameIndexSecond < framesSecond.size())
{
sampleValue += framesSecond[frameIndexSecond].Data[sampleIndexSecond++] * secondVolume;
indexSecond++;
}
}
if (sampleValue > maxVal)
maxVal = sampleValue;
if (sampleValue < minVal)
minVal = sampleValue;
mixedData.push_back(sampleValue);
if (clipToMainAudio && indexFirst >= totalSamplesFirst)
break;
}
if (maxVal > 0.75f || minVal < -0.75f)
{
float maxMul = max(maxVal, -minVal);
for (auto& fm : mixedData)
{
fm /= maxMul;
}
}
std::vector<MixedFrame> result;
size_t samplesPerFrame = framesFirst[0].Data.size();
size_t totalMixedSamples = mixedData.size();
size_t currentSampleIndex = 0;
size_t currFrameIndex = 0;
while (currentSampleIndex < totalMixedSamples)
{
if (currFrameIndex < framesFirst.size())
samplesPerFrame = framesFirst[currFrameIndex++].Data.size();
MixedFrame frame;
size_t remainingSamples = totalMixedSamples - currentSampleIndex;
size_t currentFrameSamples = min(samplesPerFrame, remainingSamples);
frame.Data.insert(frame.Data.end(),
mixedData.begin() + currentSampleIndex,
mixedData.begin() + currentSampleIndex + currentFrameSamples);
frame.Duration = static_cast<LONGLONG>((currentFrameSamples * 10000000) / samplesPerSecond);
frame.Time = static_cast<LONGLONG>((currentSampleIndex * 10000000) / samplesPerSecond);
result.push_back(frame);
currentSampleIndex += currentFrameSamples;
}
return result;
}
/*
将一个主音频和多个连续的副音频按权重进行混合
mainVolume:主音频音量权重[0-1]
secondVolume:副音频音量权重[0-1]
clipToMainAudio:是否将副音频剪切到匹配主音频的长度
loopMode:如果副音频长度比主音频短则循环副音频
*/
std::vector<MixedFrame> AudioMix::MixAudio(double samplesPerSecond, std::vector<MixedFrame> mainAudio, float mainVolume, std::vector<MixedFrame> framesSecond, float secondVolume, bool clipToMainAudio, bool loopMode)
{
size_t totalSamplesFirst = 0;
for (const auto& frame : mainAudio)
{
totalSamplesFirst += frame.Data.size();
}
size_t totalSamplesSecond = 0;
for (const auto& frame : framesSecond)
{
totalSamplesSecond += frame.Data.size();
}
size_t totalSamples = max(totalSamplesFirst, totalSamplesSecond);
std::vector<float> mixedData;
mixedData.reserve(totalSamples);
size_t indexFirst = 0, indexSecond = 0;
size_t frameIndexFirst = 0, frameIndexSecond = 0;
size_t sampleIndexFirst = 0, sampleIndexSecond = 0;
float maxVal = 0.0f, minVal = 0.0f;
while (indexFirst < totalSamplesFirst || indexSecond < totalSamplesSecond)
{
float sampleValue = 0.0f;
if (indexFirst < totalSamplesFirst)
{
if (sampleIndexFirst >= mainAudio[frameIndexFirst].Data.size())
{
frameIndexFirst++;
sampleIndexFirst = 0;
}
if (frameIndexFirst < mainAudio.size())
{
sampleValue += mainAudio[frameIndexFirst].Data[sampleIndexFirst++] * mainVolume;
indexFirst++;
}
}
if (loopMode && indexSecond >= totalSamplesSecond)
{
indexSecond = 0;
frameIndexSecond = 0;
sampleIndexSecond = 0;
}
if (indexSecond < totalSamplesSecond)
{
if (sampleIndexSecond >= framesSecond[frameIndexSecond].Data.size())
{
frameIndexSecond++;
sampleIndexSecond = 0;
}
if (frameIndexSecond < framesSecond.size())
{
sampleValue += framesSecond[frameIndexSecond].Data[sampleIndexSecond++] * secondVolume;
indexSecond++;
}
}
if (sampleValue > maxVal)
maxVal = sampleValue;
if (sampleValue < minVal)
minVal = sampleValue;
mixedData.push_back(sampleValue);
if (clipToMainAudio && indexFirst >= totalSamplesFirst)
break;
}
if (maxVal > 0.75f || minVal < -0.75f)
{
float maxMul = max(maxVal, -minVal);
for (auto& fm : mixedData)
{
fm /= maxMul;
}
}
std::vector<MixedFrame> result;
size_t samplesPerFrame = mainAudio[0].Data.size();
size_t totalMixedSamples = mixedData.size();
size_t currentSampleIndex = 0;
size_t currFrameIndex = 0;
while (currentSampleIndex < totalMixedSamples)
{
if (currFrameIndex < mainAudio.size())
samplesPerFrame = mainAudio[currFrameIndex++].Data.size();
MixedFrame frame;
size_t remainingSamples = totalMixedSamples - currentSampleIndex;
size_t currentFrameSamples = min(samplesPerFrame, remainingSamples);
frame.Data.insert(frame.Data.end(),
mixedData.begin() + currentSampleIndex,
mixedData.begin() + currentSampleIndex + currentFrameSamples);
frame.Duration = static_cast<LONGLONG>((currentFrameSamples * 10000000) / samplesPerSecond);
frame.Time = static_cast<LONGLONG>((currentSampleIndex * 10000000) / samplesPerSecond);
result.push_back(frame);
currentSampleIndex += currentFrameSamples;
}
return result;
}
H264Encoder::H264Encoder(std::wstring destFilename, UINT32 width, UINT32 height, UINT32 fps, UINT32 videoBitrate)
{
configWMFLib();
m_VideoBitrate = videoBitrate;
m_DestFilename = destFilename;
m_Width = width;
m_Height = height;
m_cbWidth = 4 * m_Width;
m_cbBuffer = m_cbWidth * m_Height;
m_VideoFPS = fps;
m_FrameDuration = (10 * 1000 * 1000 / m_VideoFPS);
MFCreateMemoryBuffer(m_cbBuffer, &m_pBuffer);
m_pBuffer->SetCurrentLength(m_cbBuffer);
InitializeWriter();
m_FrameIndex = 0;
}
H264Encoder::~H264Encoder()
{
if (this->m_pSinkWriter)
this->m_pSinkWriter->Release();
if (this->m_pBuffer)
m_pBuffer->Release();
if (this->pVideoSample)
pVideoSample->Release();
}
HRESULT H264Encoder::AddVideoFrame(IWICBitmap* buffer, LONGLONG sampleTime, LONGLONG duration)
{
if (pVideoSample == NULL)
{
MFCreateSample(&pVideoSample);
pVideoSample->AddBuffer(m_pBuffer);
}
BYTE* pData = NULL;
HRESULT hr = m_pBuffer->Lock(&pData, NULL, NULL);
if (SUCCEEDED(hr))
{
buffer->CopyPixels(NULL, m_cbWidth, m_cbWidth * m_Height, (BYTE*)pData);
}
if (m_pBuffer)
m_pBuffer->Unlock();
if (SUCCEEDED(hr))
hr = pVideoSample->SetSampleTime(sampleTime == 0 ? m_FrameDuration * m_FrameIndex : sampleTime);
if (SUCCEEDED(hr))
hr = pVideoSample->SetSampleDuration(duration == 0 ? m_FrameDuration : duration);
if (SUCCEEDED(hr))
hr = m_pSinkWriter->WriteSample(VideoStreamIndex, pVideoSample);
VideoTimeStamp += m_FrameDuration;
m_FrameIndex += 1;
return hr;
}
HRESULT H264Encoder::AddVideoFrame(BYTE* buffer, LONGLONG sampleTime, LONGLONG duration)
{
if (pVideoSample == NULL)
{
MFCreateSample(&pVideoSample);
pVideoSample->AddBuffer(m_pBuffer);
}
BYTE* pData = NULL;
HRESULT hr = m_pBuffer->Lock(&pData, NULL, NULL);
if (SUCCEEDED(hr))
{
RtlCopyMemory(pData, buffer, m_cbWidth * m_Height);
}
if (m_pBuffer)
m_pBuffer->Unlock();
if (SUCCEEDED(hr))
hr = pVideoSample->SetSampleTime(sampleTime == 0 ? m_FrameDuration * m_FrameIndex : sampleTime);
if (SUCCEEDED(hr))
hr = pVideoSample->SetSampleDuration(duration == 0 ? m_FrameDuration : duration);
if (SUCCEEDED(hr))
hr = m_pSinkWriter->WriteSample(VideoStreamIndex, pVideoSample);
VideoTimeStamp += m_FrameDuration;
m_FrameIndex += 1;
return hr;
}
HRESULT H264Encoder::AddAudioFrame(PVOID buffer, DWORD bufferSize, LONGLONG frameDuration)
{
IMFSample* pSample = NULL;
IMFMediaBuffer* m_pBuffer = NULL;
BYTE* pData = NULL;
HRESULT hr = MFCreateSample(&pSample);
if (SUCCEEDED(hr))
{
hr = MFCreateMemoryBuffer(bufferSize, &m_pBuffer);
hr = m_pBuffer->SetCurrentLength(bufferSize);
if (SUCCEEDED(hr))
{
hr = m_pBuffer->Lock(&pData, NULL, NULL);
if (SUCCEEDED(hr))
{
memcpy(pData, buffer, bufferSize);
m_pBuffer->Unlock();
}
do {
hr = pSample->AddBuffer(m_pBuffer);
BREAK_ON_FAIL(hr);
hr = pSample->SetSampleTime(this->AudioTimeStamp);
BREAK_ON_FAIL(hr);
hr = pSample->SetSampleDuration(frameDuration);
BREAK_ON_FAIL(hr);
hr = m_pSinkWriter->WriteSample(AudioStreamIndex, pSample);
BREAK_ON_FAIL(hr);
AudioTimeStamp += frameDuration;
} while (false);
m_pBuffer->Release();
}
pSample->Release();
}
return hr;
}
HRESULT H264Encoder::AddAudioFrame(IMFSample* pSample)
{
DWORD flags = 0;
HRESULT hr = S_OK;
do
{
if (pSample)
{
pSample->GetSampleFlags(&flags);
pSample->SetSampleTime(this->AudioTimeStamp);
}
else
{
AudioTimeStamp = VideoTimeStamp;
return m_pSinkWriter->SendStreamTick(AudioStreamIndex, AudioTimeStamp);
}
if (flags & MF_SOURCE_READERF_ENDOFSTREAM)
{
hr = m_pSinkWriter->SendStreamTick(AudioStreamIndex, this->GetTimeStamp());
AudioTimeStamp = this->GetTimeStamp();
BREAK_ON_FAIL(hr);
hr = MF_SOURCE_READERF_ENDOFSTREAM;
}
else
{
hr = m_pSinkWriter->WriteSample(AudioStreamIndex, pSample);
LONGLONG sampleDuration = 0;
hr = pSample->GetSampleDuration(&sampleDuration);
AudioTimeStamp += sampleDuration;
BREAK_ON_FAIL(hr);
}
pSample = NULL;
} while (false);
return hr;
}
HRESULT H264Encoder::TickAudioFrame(LONGLONG sampleTime)
{
this->AudioTimeStamp = sampleTime;
return m_pSinkWriter->SendStreamTick(AudioStreamIndex, this->AudioTimeStamp);
}
HRESULT H264Encoder::EndAudio()
{
return m_pSinkWriter->NotifyEndOfSegment(AudioStreamIndex);
}
HRESULT H264Encoder::End()
{
return m_pSinkWriter->Finalize();
}
HRESULT H264Encoder::ConfigAudioStream()
{
IMFMediaType* pMediaType = NULL;
HRESULT hr = MFCreateMediaType(&pMediaType);
do
{
BREAK_ON_NULL(pMediaType, E_POINTER);
hr = pMediaType->DeleteAllItems();
BREAK_ON_FAIL(hr);
pMediaType->SetGUID(MF_MT_MAJOR_TYPE, MFMediaType_Audio);
BREAK_ON_FAIL(hr);
hr = pMediaType->SetGUID(MF_MT_SUBTYPE, MFAudioFormat_AAC);
BREAK_ON_FAIL(hr);
hr = pMediaType->SetUINT32(MF_MT_AUDIO_BITS_PER_SAMPLE, 16);
BREAK_ON_FAIL(hr);
hr = pMediaType->SetUINT32(MF_MT_AUDIO_SAMPLES_PER_SECOND, 44100);
BREAK_ON_FAIL(hr);
hr = pMediaType->SetUINT32(MF_MT_AUDIO_NUM_CHANNELS, 2);
BREAK_ON_FAIL(hr);
hr = pMediaType->SetUINT32(MF_MT_AUDIO_AVG_BYTES_PER_SECOND, 16000);
BREAK_ON_FAIL(hr);
hr = pMediaType->SetUINT32(MF_MT_AUDIO_BLOCK_ALIGNMENT, 1);
BREAK_ON_FAIL(hr);
hr = m_pSinkWriter->AddStream(pMediaType, &AudioStreamIndex);
BREAK_ON_FAIL(hr);
pMediaType->Release();
} while (false);
return hr;
}
HRESULT H264Encoder::SetCurrentAudioType(IMFMediaType* mediaType)
{
return m_pSinkWriter->SetInputMediaType(AudioStreamIndex, mediaType, NULL);
}
HRESULT H264Encoder::InitializeWriter()
{
HRESULT hr = S_OK;
m_pSinkWriter = NULL;
VideoStreamIndex = 0;
IMFMediaType* pMediaTypeOut = nullptr;
IMFMediaType* pMediaTypeIn = nullptr;
do
{
hr = MFCreateSinkWriterFromURL(m_DestFilename.c_str(), nullptr, nullptr, &m_pSinkWriter);
BREAK_ON_FAIL(hr);
hr = SetVideoOutputType(&pMediaTypeOut);
BREAK_ON_FAIL(hr);
hr = SetVideoInputType(&pMediaTypeIn);
BREAK_ON_FAIL(hr);
hr = ConfigAudioStream();
BREAK_ON_FAIL(hr);
} while (false);
if (pMediaTypeOut) pMediaTypeOut->Release();
if (pMediaTypeIn) pMediaTypeIn->Release();
return hr;
}
HRESULT H264Encoder::Begin()
{
return m_pSinkWriter->BeginWriting();
}
HRESULT H264Encoder::SetVideoOutputType(IMFMediaType** pMediaTypeOut)
{
HRESULT hr = S_OK;
do
{
hr = MFCreateMediaType(pMediaTypeOut);
BREAK_ON_FAIL(hr);
hr = (*pMediaTypeOut)->SetGUID(MF_MT_MAJOR_TYPE, MFMediaType_Video);
BREAK_ON_FAIL(hr);
hr = (*pMediaTypeOut)->SetGUID(MF_MT_SUBTYPE, MFVideoFormat_H264);
BREAK_ON_FAIL(hr);
hr = (*pMediaTypeOut)->SetUINT32(MF_MT_MPEG2_PROFILE, eAVEncH264VProfile_High);
BREAK_ON_FAIL(hr);
hr = (*pMediaTypeOut)->SetUINT32(MF_MT_AVG_BITRATE, m_VideoBitrate);
BREAK_ON_FAIL(hr);
hr = (*pMediaTypeOut)->SetUINT32(MF_MT_INTERLACE_MODE, MFVideoInterlace_Progressive);
BREAK_ON_FAIL(hr);
hr = MFSetAttributeSize((*pMediaTypeOut), MF_MT_FRAME_SIZE, m_Width, m_Height);
BREAK_ON_FAIL(hr);
hr = MFSetAttributeRatio((*pMediaTypeOut), MF_MT_FRAME_RATE, m_VideoFPS, 1);
BREAK_ON_FAIL(hr);
hr = MFSetAttributeRatio((*pMediaTypeOut), MF_MT_PIXEL_ASPECT_RATIO, 1, 1);
BREAK_ON_FAIL(hr);
hr = m_pSinkWriter->AddStream(*pMediaTypeOut, &VideoStreamIndex);
} while (false);
return hr;
}
HRESULT H264Encoder::SetVideoInputType(IMFMediaType** pMediaTypeIn)
{
HRESULT hr = S_OK;
do
{
hr = MFCreateMediaType(pMediaTypeIn);
BREAK_ON_FAIL(hr);
hr = (*pMediaTypeIn)->SetGUID(MF_MT_MAJOR_TYPE, MFMediaType_Video);
BREAK_ON_FAIL(hr);
hr = (*pMediaTypeIn)->SetGUID(MF_MT_SUBTYPE, MFVideoFormat_RGB32);
BREAK_ON_FAIL(hr);
hr = (*pMediaTypeIn)->SetUINT32(MF_MT_INTERLACE_MODE, MFVideoInterlace_Progressive);
BREAK_ON_FAIL(hr);
hr = MFSetAttributeSize(*pMediaTypeIn, MF_MT_FRAME_SIZE, m_Width, m_Height);
BREAK_ON_FAIL(hr);
hr = MFSetAttributeRatio(*pMediaTypeIn, MF_MT_FRAME_RATE, m_VideoFPS, 1);
BREAK_ON_FAIL(hr);
hr = MFSetAttributeRatio(*pMediaTypeIn, MF_MT_PIXEL_ASPECT_RATIO, 1, 1);
BREAK_ON_FAIL(hr);
hr = m_pSinkWriter->SetInputMediaType(VideoStreamIndex, *pMediaTypeIn, NULL);
} while (false);
return hr;
}
HRESULT H264Encoder::InitTranscodeVideoType(CComPtr<IMFMediaType>& pStreamMediaType)
{
HRESULT hr = S_OK;
do
{
BREAK_ON_NULL(pStreamMediaType, E_POINTER);
hr = pStreamMediaType->DeleteAllItems();
BREAK_ON_FAIL(hr);
pStreamMediaType->SetGUID(MF_MT_MAJOR_TYPE, MFMediaType_Video);
BREAK_ON_FAIL(hr);
hr = pStreamMediaType->SetGUID(MF_MT_SUBTYPE, MFVideoFormat_H264);
BREAK_ON_FAIL(hr);
hr = MFSetAttributeSize(
pStreamMediaType, MF_MT_FRAME_SIZE, m_Width, m_Height);
BREAK_ON_FAIL(hr);
hr = MFSetAttributeRatio(
pStreamMediaType, MF_MT_FRAME_RATE, m_VideoFPS, 1);
BREAK_ON_FAIL(hr);
hr = pStreamMediaType->SetUINT32(MF_MT_AVG_BITRATE, m_VideoBitrate);
BREAK_ON_FAIL(hr);
hr = pStreamMediaType->SetUINT32(MF_MT_INTERLACE_MODE,
MFVideoInterlace_Progressive);
BREAK_ON_FAIL(hr);
hr = MFSetAttributeSize(pStreamMediaType, MF_MT_PIXEL_ASPECT_RATIO, 1, 1);
BREAK_ON_FAIL(hr);
} while (false);
return hr;
}
double H264Encoder::CurrentTimeOfSeconds()
{
return double(m_FrameIndex) / double(m_VideoFPS);
}
UINT64 H264Encoder::GetTimeStamp()
{
return m_FrameDuration * m_FrameIndex;
}
UINT64 H264Encoder::GetTimeStampForIndex(UINT32 index)
{
return m_FrameDuration * index;
}
AudioCapture::AudioCapture()
{
configWMFLib();
HRESULT hr = CoCreateInstance(
__uuidof(MMDeviceEnumerator),
NULL, CLSCTX_ALL,
__uuidof(IMMDeviceEnumerator),
(void**)&pEnumerator
);
hr = pEnumerator->GetDefaultAudioEndpoint(eRender, eConsole, &pDevice);
hr = pDevice->Activate(
__uuidof(IAudioClient),
CLSCTX_ALL,
NULL,
(void**)&pAudioClient
);
hr = pAudioClient->GetMixFormat(&pwfx);
hr = pAudioClient->Initialize(
AUDCLNT_SHAREMODE_SHARED,
AUDCLNT_STREAMFLAGS_LOOPBACK,
0, 0, pwfx, NULL
);
hr = pAudioClient->GetService(
__uuidof(IAudioCaptureClient),
(void**)&pCaptureClient
);
}
AudioCapture::~AudioCapture()
{
CoTaskMemFree(pwfx);
pCaptureClient->Release();
pAudioClient->Release();
pDevice->Release();
pEnumerator->Release();
}
HRESULT AudioCapture::Start()
{
return pAudioClient->Start();
}
HRESULT AudioCapture::Stop()
{
return pAudioClient->Stop();
}
std::vector<uint8_t> AudioCapture::Capture(UINT32* packetLength, UINT32* numFramesAvailable, DWORD* dwFlags)
{
BYTE* pData = NULL;
std::vector<uint8_t> result;
if (SUCCEEDED(pCaptureClient->GetNextPacketSize(packetLength)) && *packetLength > 0)
{
HRESULT hr = pCaptureClient->GetBuffer(&pData, numFramesAvailable, dwFlags, NULL, NULL);
if (SUCCEEDED(hr))
{
UINT32 bufferSize = *numFramesAvailable * pwfx->nBlockAlign;
if (bufferSize == 0)
{
pCaptureClient->ReleaseBuffer(*numFramesAvailable);
return result;
}
result.resize(bufferSize);
memcpy(result.data(), pData, bufferSize);
pCaptureClient->ReleaseBuffer(*numFramesAvailable);
}
}
return result;
}
HRESULT AudioCapture::Capture(BYTE* buffer, UINT32* bufferSize, LONGLONG* duration)
{
HRESULT hr = S_OK;
BYTE* pData = NULL;
UINT32 packetLength;
UINT32 numFramesAvailable;
DWORD dwFlags;
*bufferSize = 0;
*duration = 0;
hr = pCaptureClient->GetNextPacketSize(&packetLength);
if (FAILED(hr))
{
return hr;
}
if (packetLength <= 0)
{
return AUDCLNT_S_BUFFER_EMPTY;
}
hr = pCaptureClient->GetBuffer(&pData, &numFramesAvailable, &dwFlags, NULL, NULL);
if (FAILED(hr) || hr == AUDCLNT_S_BUFFER_EMPTY)
{
return hr;
}
*bufferSize = numFramesAvailable * pwfx->nBlockAlign;
if (*bufferSize == 0)
{
pCaptureClient->ReleaseBuffer(numFramesAvailable);
return AUDCLNT_S_BUFFER_EMPTY;
}
memcpy(buffer, pData, *bufferSize);
*duration = ((double)numFramesAvailable / (double)pwfx->nSamplesPerSec) * 10000000.0;
pCaptureClient->ReleaseBuffer(numFramesAvailable);
return S_OK;
}
UINT32 AudioCapture::BytesPerFrame()
{
return pwfx->nBlockAlign;
}
DWORD AudioCapture::Channels()
{
return pwfx->nChannels;
}
DWORD AudioCapture::SamplesPerSecond()
{
return pwfx->nSamplesPerSec;
}
IMFMediaType* AudioCapture::GetMediaType()
{
HRESULT hr = S_OK;
IMFMediaType* pMediaTypeOut = NULL;
hr = MFCreateMediaType(&pMediaTypeOut);
if (SUCCEEDED(hr)) {
hr = pMediaTypeOut->SetGUID(MF_MT_MAJOR_TYPE, MFMediaType_Audio);
}
if (SUCCEEDED(hr)) {
hr = pMediaTypeOut->SetGUID(MF_MT_SUBTYPE, MFAudioFormat_Float);
}
if (SUCCEEDED(hr)) {
hr = pMediaTypeOut->SetUINT32(MF_MT_AUDIO_NUM_CHANNELS, pwfx->nChannels);
}
if (SUCCEEDED(hr)) {
hr = pMediaTypeOut->SetUINT32(MF_MT_AUDIO_SAMPLES_PER_SECOND, pwfx->nSamplesPerSec);
}
if (SUCCEEDED(hr)) {
hr = pMediaTypeOut->SetUINT32(MF_MT_AUDIO_BLOCK_ALIGNMENT, pwfx->nBlockAlign);
}
if (SUCCEEDED(hr)) {
hr = pMediaTypeOut->SetUINT32(MF_MT_AUDIO_AVG_BYTES_PER_SECOND, pwfx->nAvgBytesPerSec);
}
if (SUCCEEDED(hr)) {
hr = pMediaTypeOut->SetUINT32(MF_MT_AUDIO_BITS_PER_SAMPLE, pwfx->wBitsPerSample);
}
if (SUCCEEDED(hr)) {
hr = pMediaTypeOut->SetUINT32(MF_MT_ALL_SAMPLES_INDEPENDENT, TRUE);
}
return pMediaTypeOut;
}
Reference in New Issue View Git Blame Copy Permalink