mirror of
https://github.com/kunkundi/crossdesk.git
synced 2025-10-27 04:35:34 +08:00
[feat] support speaker capture on Windows
This commit is contained in:
dijunkun
@@ -1,187 +1,94 @@
|
||||
#include "speaker_capturer_wasapi.h"
|
||||
|
||||
#include <algorithm>
|
||||
#include <climits>
|
||||
#include <iostream>
|
||||
#include "rd_log.h"
|
||||
|
||||
#define REFTIMES_PER_SEC 10000000
|
||||
#define REFTIMES_PER_MILLISEC 10000
|
||||
#define MINIAUDIO_IMPLEMENTATION
|
||||
#include "miniaudio.h"
|
||||
|
||||
#define SAVE_AUDIO_FILE 0
|
||||
|
||||
#define CHECK_HR(hres) \
|
||||
if (FAILED(hres)) { \
|
||||
return -1; \
|
||||
static ma_device_config device_config_;
|
||||
static ma_device device_;
|
||||
static ma_format format_ = ma_format_s16;
|
||||
static ma_uint32 sample_rate_ = ma_standard_sample_rate_48000;
|
||||
static ma_uint32 channels_ = 1;
|
||||
static FILE* fp_ = nullptr;
|
||||
|
||||
void data_callback(ma_device* pDevice, void* pOutput, const void* pInput,
|
||||
ma_uint32 frameCount) {
|
||||
SpeakerCapturerWasapi* ptr = (SpeakerCapturerWasapi*)pDevice->pUserData;
|
||||
if (ptr) {
|
||||
if (SAVE_AUDIO_FILE) {
|
||||
fwrite(pInput, frameCount * ma_get_bytes_per_frame(format_, channels_), 1,
|
||||
fp_);
|
||||
}
|
||||
|
||||
ptr->GetCallback()((unsigned char*)pInput,
|
||||
frameCount * ma_get_bytes_per_frame(format_, channels_));
|
||||
}
|
||||
|
||||
#define SAFE_RELEASE(punk) \
|
||||
if ((punk) != nullptr) { \
|
||||
(punk)->Release(); \
|
||||
(punk) = nullptr; \
|
||||
}
|
||||
(void)pOutput;
|
||||
}
|
||||
|
||||
const CLSID CLSID_MMDeviceEnumerator = __uuidof(MMDeviceEnumerator);
|
||||
const IID IID_IMMDeviceEnumerator = __uuidof(IMMDeviceEnumerator);
|
||||
const IID IID_IAudioClient = __uuidof(IAudioClient);
|
||||
const IID IID_IAudioCaptureClient = __uuidof(IAudioCaptureClient);
|
||||
SpeakerCapturerWasapi::speaker_data_cb SpeakerCapturerWasapi::GetCallback() {
|
||||
return cb_;
|
||||
}
|
||||
|
||||
SpeakerCapturerWasapi::SpeakerCapturerWasapi() {}
|
||||
|
||||
SpeakerCapturerWasapi::~SpeakerCapturerWasapi() {
|
||||
if (inited_ && capture_thread_->joinable()) {
|
||||
capture_thread_->join();
|
||||
inited_ = false;
|
||||
}
|
||||
|
||||
CoTaskMemFree(pwfx);
|
||||
SAFE_RELEASE(pEnumerator)
|
||||
SAFE_RELEASE(pDevice)
|
||||
SAFE_RELEASE(pAudioClient)
|
||||
SAFE_RELEASE(pCaptureClient)
|
||||
|
||||
if (SAVE_AUDIO_FILE) {
|
||||
fclose(fp);
|
||||
fclose(fp_);
|
||||
}
|
||||
|
||||
// if (pData_dst) delete pData_dst;
|
||||
// pData_dst = nullptr;
|
||||
}
|
||||
|
||||
int SpeakerCapturerWasapi::Init(speaker_data_cb cb) {
|
||||
if (inited_) {
|
||||
return 0;
|
||||
}
|
||||
|
||||
cb_ = cb;
|
||||
|
||||
if (SAVE_AUDIO_FILE) {
|
||||
fopen_s(&fp, "system_audio.pcm", "wb");
|
||||
fopen_s(&fp_, "system_audio.pcm", "wb");
|
||||
}
|
||||
|
||||
HRESULT hr;
|
||||
ma_result result;
|
||||
ma_backend backends[] = {ma_backend_wasapi};
|
||||
|
||||
hr = CoCreateInstance(CLSID_MMDeviceEnumerator, nullptr, CLSCTX_ALL,
|
||||
IID_IMMDeviceEnumerator, (void **)&pEnumerator);
|
||||
CHECK_HR(hr)
|
||||
device_config_ = ma_device_config_init(ma_device_type_loopback);
|
||||
device_config_.capture.pDeviceID = NULL;
|
||||
device_config_.capture.format = format_;
|
||||
device_config_.capture.channels = channels_;
|
||||
device_config_.sampleRate = sample_rate_;
|
||||
device_config_.dataCallback = data_callback;
|
||||
device_config_.pUserData = this;
|
||||
|
||||
hr = pEnumerator->GetDefaultAudioEndpoint(eRender, eConsole,
|
||||
&pDevice); // 输出
|
||||
CHECK_HR(hr)
|
||||
|
||||
hr = pDevice->Activate(IID_IAudioClient, CLSCTX_ALL, nullptr,
|
||||
(void **)&pAudioClient);
|
||||
CHECK_HR(hr)
|
||||
|
||||
hr = pAudioClient->GetMixFormat(&pwfx);
|
||||
CHECK_HR(hr)
|
||||
|
||||
// Change to 16bit
|
||||
if (pwfx->wFormatTag == WAVE_FORMAT_IEEE_FLOAT) {
|
||||
pwfx->wFormatTag = WAVE_FORMAT_PCM;
|
||||
pwfx->wBitsPerSample = 16;
|
||||
pwfx->nBlockAlign = pwfx->nChannels * pwfx->wBitsPerSample / 8;
|
||||
pwfx->nAvgBytesPerSec = pwfx->nBlockAlign * pwfx->nSamplesPerSec;
|
||||
} else if (pwfx->wFormatTag == WAVE_FORMAT_EXTENSIBLE) {
|
||||
PWAVEFORMATEXTENSIBLE pEx = reinterpret_cast<PWAVEFORMATEXTENSIBLE>(pwfx);
|
||||
if (IsEqualGUID(KSDATAFORMAT_SUBTYPE_IEEE_FLOAT, pEx->SubFormat)) {
|
||||
pEx->SubFormat = KSDATAFORMAT_SUBTYPE_PCM;
|
||||
pEx->Samples.wValidBitsPerSample = 16;
|
||||
pwfx->wBitsPerSample = 16;
|
||||
pwfx->nBlockAlign = pwfx->nChannels * pwfx->wBitsPerSample / 8;
|
||||
pwfx->nAvgBytesPerSec = pwfx->nBlockAlign * pwfx->nSamplesPerSec;
|
||||
}
|
||||
result = ma_device_init_ex(backends, sizeof(backends) / sizeof(backends[0]),
|
||||
NULL, &device_config_, &device_);
|
||||
if (result != MA_SUCCESS) {
|
||||
LOG_ERROR("Failed to initialize loopback device");
|
||||
return -1;
|
||||
}
|
||||
|
||||
hr = pAudioClient->Initialize(AUDCLNT_SHAREMODE_SHARED,
|
||||
AUDCLNT_STREAMFLAGS_LOOPBACK, 0, 0, pwfx,
|
||||
nullptr);
|
||||
CHECK_HR(hr)
|
||||
|
||||
// Get the size of the allocated buffer.
|
||||
hr = pAudioClient->GetBufferSize(&bufferFrameCount);
|
||||
CHECK_HR(hr)
|
||||
|
||||
hr = pAudioClient->GetService(IID_IAudioCaptureClient,
|
||||
(void **)&pCaptureClient);
|
||||
CHECK_HR(hr)
|
||||
|
||||
// Show audio info
|
||||
{
|
||||
printf("wFormatTag is %x\n", pwfx->wFormatTag);
|
||||
printf("nChannels is %x\n", pwfx->nChannels);
|
||||
printf("nSamplesPerSec is %d\n", pwfx->nSamplesPerSec);
|
||||
printf("nAvgBytesPerSec is %d\n", pwfx->nAvgBytesPerSec);
|
||||
printf("wBitsPerSample is %d\n", pwfx->wBitsPerSample);
|
||||
}
|
||||
|
||||
hnsActualDuration =
|
||||
(double)REFTIMES_PER_SEC * bufferFrameCount / pwfx->nSamplesPerSec;
|
||||
|
||||
// pData_dst = new BYTE[960];
|
||||
|
||||
inited_ = true;
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
int SpeakerCapturerWasapi::Start() {
|
||||
HRESULT hr;
|
||||
hr = pAudioClient->Start();
|
||||
CHECK_HR(hr)
|
||||
|
||||
capture_thread_.reset(new std::thread([this]() {
|
||||
HRESULT hr;
|
||||
|
||||
// Each loop fills about half of the shared buffer.
|
||||
while (1) {
|
||||
// Sleep for half the buffer duration.
|
||||
Sleep(hnsActualDuration / REFTIMES_PER_MILLISEC / 4);
|
||||
|
||||
hr = pCaptureClient->GetNextPacketSize(&packetLength);
|
||||
CHECK_HR(hr)
|
||||
|
||||
while (packetLength != 0) {
|
||||
// Get the available data in the shared buffer.
|
||||
hr = pCaptureClient->GetBuffer(&pData, &numFramesAvailable, &flags,
|
||||
nullptr, &ts);
|
||||
CHECK_HR(hr)
|
||||
|
||||
// flags equals to 2 means silence, set data to nullptr
|
||||
if (flags == AUDCLNT_BUFFERFLAGS_SILENT) {
|
||||
pData = nullptr;
|
||||
}
|
||||
|
||||
if (pData != nullptr) {
|
||||
size_t size = numFramesAvailable * pwfx->nBlockAlign;
|
||||
|
||||
for (int i = 0; i < size / 2; i++) {
|
||||
BYTE left = pData[i * 2];
|
||||
BYTE right = pData[i * 2 + 1];
|
||||
// Right channel only?
|
||||
BYTE monoSample = right;
|
||||
|
||||
pData_dst[i] = static_cast<BYTE>(monoSample);
|
||||
}
|
||||
|
||||
cb_(pData_dst, size / 2);
|
||||
|
||||
if (SAVE_AUDIO_FILE) {
|
||||
fwrite(pData_dst, size / 2, 1, fp);
|
||||
}
|
||||
}
|
||||
|
||||
hr = pCaptureClient->ReleaseBuffer(numFramesAvailable);
|
||||
CHECK_HR(hr)
|
||||
|
||||
hr = pCaptureClient->GetNextPacketSize(&packetLength);
|
||||
CHECK_HR(hr)
|
||||
}
|
||||
}
|
||||
}));
|
||||
ma_result result = ma_device_start(&device_);
|
||||
if (result != MA_SUCCESS) {
|
||||
ma_device_uninit(&device_);
|
||||
LOG_ERROR("Failed to start device");
|
||||
return -1;
|
||||
}
|
||||
|
||||
return 0;
|
||||
}
|
||||
|
||||
int SpeakerCapturerWasapi::Stop() {
|
||||
HRESULT hr;
|
||||
hr = pAudioClient->Stop();
|
||||
CHECK_HR(hr)
|
||||
ma_device_uninit(&device_);
|
||||
return 0;
|
||||
}
|
||||
|
||||
|
||||
@@ -1,20 +1,12 @@
|
||||
/*
|
||||
* @Author: DI JUNKUN
|
||||
* @Date: 2024-07-22
|
||||
* @Date: 2024-08-15
|
||||
* Copyright (c) 2024 by DI JUNKUN, All Rights Reserved.
|
||||
*/
|
||||
|
||||
#ifndef _SPEAKER_CAPTURER_WASAPI_H_
|
||||
#define _SPEAKER_CAPTURER_WASAPI_H_
|
||||
|
||||
#include <Audioclient.h>
|
||||
#include <Devicetopology.h>
|
||||
#include <Endpointvolume.h>
|
||||
#include <Mmdeviceapi.h>
|
||||
|
||||
#include <thread>
|
||||
#include <vector>
|
||||
|
||||
#include "speaker_capturer.h"
|
||||
|
||||
class SpeakerCapturerWasapi : public SpeakerCapturer {
|
||||
@@ -31,31 +23,13 @@ class SpeakerCapturerWasapi : public SpeakerCapturer {
|
||||
int Pause();
|
||||
int Resume();
|
||||
|
||||
speaker_data_cb GetCallback();
|
||||
|
||||
private:
|
||||
speaker_data_cb cb_ = nullptr;
|
||||
|
||||
private:
|
||||
REFERENCE_TIME hnsActualDuration;
|
||||
UINT32 bufferFrameCount;
|
||||
UINT32 numFramesAvailable;
|
||||
BYTE *pData;
|
||||
// std::vector<BYTE> pData_dst;
|
||||
BYTE pData_dst[960];
|
||||
DWORD flags;
|
||||
|
||||
// REFERENCE_TIME hnsRequestedDuration = 10000000;
|
||||
IMMDeviceEnumerator *pEnumerator = NULL;
|
||||
IMMDevice *pDevice = NULL;
|
||||
IAudioClient *pAudioClient = NULL;
|
||||
IAudioCaptureClient *pCaptureClient = NULL;
|
||||
WAVEFORMATEX *pwfx = NULL;
|
||||
UINT32 packetLength = 0;
|
||||
UINT64 pos, ts;
|
||||
FILE *fp;
|
||||
|
||||
bool inited_ = false;
|
||||
// thread
|
||||
std::unique_ptr<std::thread> capture_thread_ = nullptr;
|
||||
};
|
||||
|
||||
#endif
|
||||
Reference in New Issue
Block a user