From: Bernhard Kölbl besentv@gmail.com
Signed-off-by: Bernhard Kölbl besentv@gmail.com --- dlls/windows.media.speech/recognizer.c | 120 +++++++++++++++++++++++-- 1 file changed, 115 insertions(+), 5 deletions(-)
diff --git a/dlls/windows.media.speech/recognizer.c b/dlls/windows.media.speech/recognizer.c index 4abe84b7d39..d45cf03be19 100644 --- a/dlls/windows.media.speech/recognizer.c +++ b/dlls/windows.media.speech/recognizer.c @@ -19,6 +19,10 @@
#include "private.h"
+#include "initguid.h" +#include "audioclient.h" +#include "mmdeviceapi.h" + #include "wine/debug.h"
WINE_DEFAULT_DEBUG_CHANNEL(speech); @@ -161,7 +165,11 @@ struct session struct list completed_handlers; struct list result_handlers;
- HANDLE worker_thread, worker_control_event; + IAudioClient *audio_client; + IAudioCaptureClient *capture_client; + WAVEFORMATEX capture_wfx; + + HANDLE worker_thread, worker_control_event, audio_buf_event; BOOLEAN worker_running, worker_paused; CRITICAL_SECTION cs; }; @@ -181,24 +189,64 @@ static DWORD CALLBACK session_worker_thread_cb( void *args ) { ISpeechContinuousRecognitionSession *iface = args; struct session *impl = impl_from_ISpeechContinuousRecognitionSession(iface); - BOOLEAN running = TRUE; - DWORD status; + BOOLEAN running = TRUE, paused = FALSE; + DWORD flags, status; + UINT32 frame_count; + HANDLE events[2]; + BYTE *audio_buf;
SetThreadDescription(GetCurrentThread(), L"wine_speech_recognition_session_worker");
+ IAudioClient_Start(impl->audio_client); + IAudioClient_GetBufferSize(impl->audio_client, &frame_count); + while (running) { - status = WaitForMultipleObjects(1, &impl->worker_control_event, FALSE, INFINITE); + BOOLEAN old_paused = paused; + UINT32 count = 0; + + events[count++] = impl->worker_control_event; + if (!paused) events[count++] = impl->audio_buf_event; + + status = WaitForMultipleObjects(count, events, FALSE, INFINITE); if (status == 0) /* worker_control_event signaled */ { EnterCriticalSection(&impl->cs); + paused = impl->worker_paused; running = impl->worker_running; LeaveCriticalSection(&impl->cs); + + if (old_paused < paused) + { + IAudioClient_Stop(impl->audio_client); + IAudioClient_Reset(impl->audio_client); + TRACE("session worker paused.\n"); + } + else if (old_paused > paused) + { + IAudioClient_Start(impl->audio_client); + TRACE("session worker resumed.\n"); + } } + else if (status == 1) /* audio_buf_event signaled */ + { + UINT32 frames_available = 0;
- /* TODO: Send mic data to recognizer and handle results. */ + while (IAudioCaptureClient_GetBuffer(impl->capture_client, &audio_buf, &frames_available, &flags, NULL, NULL) == S_OK) + { + /* TODO: Send mic data to recognizer and handle results. */ + IAudioCaptureClient_ReleaseBuffer(impl->capture_client, frames_available); + } + } + else + { + ERR("Unexpected state entered. Aborting worker!\n"); + break; + } }
+ IAudioClient_Stop(impl->audio_client); + return 0; }
@@ -252,6 +300,9 @@ static ULONG WINAPI session_Release( ISpeechContinuousRecognitionSession *iface typed_event_handlers_clear(&impl->completed_handlers); typed_event_handlers_clear(&impl->result_handlers);
+ IAudioCaptureClient_Release(impl->capture_client); + IAudioClient_Release(impl->audio_client); + impl->cs.DebugInfo->Spare[0] = 0; DeleteCriticalSection(&impl->cs);
@@ -926,6 +977,60 @@ static const struct IActivationFactoryVtbl activation_factory_vtbl =
DEFINE_IINSPECTABLE(recognizer_factory, ISpeechRecognizerFactory, struct recognizer_statics, IActivationFactory_iface)
+static HRESULT recognizer_factory_create_audio_capture(struct session *session) +{ + const REFERENCE_TIME buffer_duration = 5000000; /* 0.5 second */ + IMMDeviceEnumerator *mm_enum = NULL; + IMMDevice *mm_device = NULL; + WAVEFORMATEX wfx = { 0 }; + WCHAR *str = NULL; + HRESULT hr = S_OK; + + if (!(session->audio_buf_event = CreateEventW(NULL, FALSE, FALSE, NULL))) + return HRESULT_FROM_WIN32(GetLastError()); + + if (FAILED(hr = CoCreateInstance(&CLSID_MMDeviceEnumerator, NULL, CLSCTX_INPROC_SERVER, &IID_IMMDeviceEnumerator, (void **)&mm_enum))) + goto cleanup; + + if (FAILED(hr = IMMDeviceEnumerator_GetDefaultAudioEndpoint(mm_enum, eCapture, eMultimedia, &mm_device))) + goto cleanup; + + if (FAILED(hr = IMMDevice_Activate(mm_device, &IID_IAudioClient, CLSCTX_INPROC_SERVER, NULL, (void **)&session->audio_client))) + goto cleanup; + + if (SUCCEEDED(hr = IMMDevice_GetId(mm_device, &str))) + TRACE("selected capture device ID: %s\n", debugstr_w(str)); + + wfx.wFormatTag = WAVE_FORMAT_PCM; + wfx.nSamplesPerSec = 16000; + wfx.nChannels = 1; + wfx.wBitsPerSample = 16; + wfx.nBlockAlign = (wfx.wBitsPerSample + 7) / 8 * wfx.nChannels; + wfx.nAvgBytesPerSec = wfx.nSamplesPerSec * wfx.nBlockAlign; + TRACE("wfx tag %u, channels %u, samples %lu, bits %u, align %u.\n", wfx.wFormatTag, wfx.nChannels, wfx.nSamplesPerSec, wfx.wBitsPerSample, wfx.nBlockAlign); + + if (FAILED(hr = IAudioClient_Initialize(session->audio_client, AUDCLNT_SHAREMODE_SHARED, AUDCLNT_STREAMFLAGS_EVENTCALLBACK, buffer_duration, 0, &wfx, NULL))) + goto cleanup; + + if (FAILED(hr = IAudioClient_SetEventHandle(session->audio_client, session->audio_buf_event))) + goto cleanup; + + hr = IAudioClient_GetService(session->audio_client, &IID_IAudioCaptureClient, (void **)&session->capture_client); + + session->capture_wfx = wfx; + +cleanup: + if (FAILED(hr)) + { + if (session->audio_client) IAudioClient_Release(session->audio_client); + if (session->audio_buf_event) CloseHandle(session->audio_buf_event); + } + if (mm_device) IMMDevice_Release(mm_device); + if (mm_enum) IMMDeviceEnumerator_Release(mm_enum); + CoTaskMemFree(str); + return hr; +} + static HRESULT WINAPI recognizer_factory_Create( ISpeechRecognizerFactory *iface, ILanguage *language, ISpeechRecognizer **speechrecognizer ) { struct recognizer *impl; @@ -969,6 +1074,9 @@ static HRESULT WINAPI recognizer_factory_Create( ISpeechRecognizerFactory *iface if (FAILED(hr = vector_inspectable_create(&constraints_iids, (IVector_IInspectable**)&session->constraints))) goto error;
+ if (FAILED(hr = recognizer_factory_create_audio_capture(session))) + goto error; + InitializeCriticalSection(&session->cs); session->cs.DebugInfo->Spare[0] = (DWORD_PTR)(__FILE__ ": recognition_session.cs");
@@ -985,6 +1093,8 @@ static HRESULT WINAPI recognizer_factory_Create( ISpeechRecognizerFactory *iface return S_OK;
error: + if (session->capture_client) IAudioCaptureClient_Release(session->capture_client); + if (session->audio_client) IAudioClient_Release(session->audio_client); if (session->constraints) IVector_ISpeechRecognitionConstraint_Release(session->constraints); CloseHandle(session->worker_control_event); free(session);