-
Notifications
You must be signed in to change notification settings - Fork 0
/
Source.cpp
323 lines (236 loc) · 9.61 KB
/
Source.cpp
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
#include <iostream>
#include <cstdio>
#include <filesystem>
#include <fstream>
#include <string>
#include <windows.h>
#include <mmeapi.h>
#include <mmdeviceapi.h>
#include <Audioclient.h>
#include <Functiondiscoverykeys_devpkey.h>
#include <Roapi.h>
#include <endpointvolume.h>
#include <ksmedia.h>
#include "whisper.h"
using namespace std;
#define REFTIMES_PER_SEC 10000000
#define RELEASE(pointer) \
if(pointer != NULL) \
pointer->Release();
class BufferSamples {
int sampleState;
public:
BufferSamples() : sampleState{ 0 } {}
std::vector<float> buf;
void push_samples(const float* samples, int sampleCount, bool dualChannel = true) {
for (int i = 0; i < sampleCount; i++) {
if (!sampleState) buf.push_back(*samples);
sampleState = (sampleState + 1) % 3;
if (dualChannel) samples += 2;
else samples += 1;
}
}
}samples;
class WavFile {
std::filesystem::path filename;
std::ofstream file;
UINT32 channels;
UINT32 sampleSize;
std::ofstream::pos_type sizePos;
std::ofstream::pos_type factChunkSampleLengthPos;
std::ofstream::pos_type dataSizePos;
public:
WavFile(const char* filename) : filename(filename) {
file = std::ofstream(this->filename.c_str(), std::ios_base::out | std::ios_base::binary);
file.write("RIFF", 4);
sizePos = file.tellp();
char size[] = { 0x48, 0x00, 0x00, 0x00 };
file.write(size, 4);
file.write("WAVE", 4);
file.write("fmt ", 4);
char sizeFmtChunk[] = { 0x28, 0x00, 0x00, 0x00 };
file.write(sizeFmtChunk, 4);
}
void setChannels(UINT32 inChannels) {
this->channels = inChannels;
}
void setSampleSize(UINT32 inSampleSize) {
this->sampleSize = inSampleSize;
}
void writeStruct(const WAVEFORMATEXTENSIBLE* waveFormatExtensible) {
file.write((char*)waveFormatExtensible, sizeof(WAVEFORMATEXTENSIBLE));
//writing the waveformatextensible file https://www.mmsp.ece.mcgill.ca/Documents/AudioFormats/WAVE/WAVE.html
file.write("fact", 4);
char sizeFactChunk[] = { 0x4, 0, 0, 0 };
file.write(sizeFactChunk, 4);
char factChunkSampleLength[] = { 0, 0, 0, 0 };
factChunkSampleLengthPos = file.tellp();
file.write(factChunkSampleLength, 4);
file.write("data", 4);
char dataSize[] = { 0,0, 0, 0 };
dataSizePos = file.tellp();
file.write(dataSize, 4);
}
void writeData(BYTE* data, UINT32 dataSize) {
file.write((const char*)data, dataSize);
}
void endData(UINT32 numOfSamples) {
file.seekp(sizePos);
UINT32 totalSize = 72 + numOfSamples * channels * sampleSize;
file.write((const char*)&totalSize, 4);
file.seekp(factChunkSampleLengthPos);
UINT32 sampleLength = channels * numOfSamples;
file.write((const char*)&sampleLength, 4);
file.seekp(dataSizePos);
UINT32 sampledDataSize = totalSize - 72;
file.write((const char*)&sampledDataSize, 4);
}
~WavFile() {
file.close();
}
};
#define EXIT_ERROR(hr) \
if(FAILED(hr)){ \
std::cout<<std::hex<<hr<<'\n'; \
goto EXIT; \
}
class AudioCap :public WavFile, BufferSamples {
const CLSID local_CLSID_MMDeviceEnumerator = __uuidof(MMDeviceEnumerator);
//interface ID
const IID local_IID_IMMDeviceEnumerator = __uuidof(IMMDeviceEnumerator);
//enumerator interface object
IMMDeviceEnumerator* pEnumerator = NULL;
//device collection
IMMDevice* device = NULL;
//property structures
IPropertyStore* properties = NULL;
PROPVARIANT pv;
IAudioClient* audioClient = NULL;
WAVEFORMATEX* waveFormat = NULL;
IAudioCaptureClient* audioCaptureClient = NULL;
IID local_IID_IAudioCaptureClient = __uuidof(IAudioCaptureClient);
IID local_IID_IAudioClient = __uuidof(IAudioClient);
UINT32 numOfBufferFrames;
REFERENCE_TIME actualDuration;
bool done = FALSE;
UINT32 framesRead = 0;
//misc variables
UINT32 bytesPerSample = 0;
UINT32 channels = 0;
UINT32 samplesPerSec = 0;
IAudioEndpointVolume* endpointvol;
float vol;
float newvol = 1.0;
HRESULT hr;
vector<float>buffer_txt;
public:
AudioCap() : WavFile("recorded.wav") {
struct whisper_context* ctx;
WHISPER_API struct whisper_full_params wparams;
int n_segments;
hr = Windows::Foundation::Initialize(RO_INIT_MULTITHREADED);
EXIT_ERROR(hr)
hr = CoCreateInstance(local_CLSID_MMDeviceEnumerator,
NULL, CLSCTX_ALL, local_IID_IMMDeviceEnumerator, (void**)&pEnumerator);
EXIT_ERROR(hr)
hr = pEnumerator->GetDefaultAudioEndpoint(eCapture, eCommunications, &device);
EXIT_ERROR(hr)
hr = device->OpenPropertyStore(STGM_READ, &properties);
EXIT_ERROR(hr)
hr = properties->GetValue(PKEY_Device_FriendlyName, &pv);
EXIT_ERROR(hr);
if (pv.vt == VT_EMPTY)
return;
else {
std::printf("End point device used : %ls\n", pv.pwszVal);
}
device->Activate(local_IID_IAudioClient, CLSCTX_ALL, NULL, (void**)&audioClient);
hr = audioClient->GetMixFormat(&waveFormat);
EXIT_ERROR(hr)
if (waveFormat->wFormatTag == WAVE_FORMAT_EXTENSIBLE) {
std::cout << "Got Extensible data structure for wave format\n";
PWAVEFORMATEXTENSIBLE waveFormatExtensible = (PWAVEFORMATEXTENSIBLE)waveFormat;
if (waveFormatExtensible->SubFormat == KSDATAFORMAT_SUBTYPE_IEEE_FLOAT) {
std::cout << "The subtype for Extensible structure is WAVE_FORMAT_FLOAT\n";
std::printf("The valid bits per sample are: %d\n", waveFormatExtensible->Samples.wValidBitsPerSample);
bytesPerSample = waveFormatExtensible->Format.wBitsPerSample / 8;
channels = waveFormatExtensible->Format.nChannels;
samplesPerSec = waveFormatExtensible->Format.nSamplesPerSec;
}
else {
EXIT_ERROR(hr)
}
}
else {
std::cout << "Default end point shared mode format is unsupported. wFormatTag = " << waveFormat->wFormatTag << '\n';
EXIT_ERROR(hr)
}
writeStruct((WAVEFORMATEXTENSIBLE*)waveFormat);
setChannels(channels);
setSampleSize(bytesPerSample);
hr = audioClient->Initialize(AUDCLNT_SHAREMODE_SHARED, 0,
REFTIMES_PER_SEC, 0, waveFormat, NULL);
EXIT_ERROR(hr)
hr = audioClient->GetBufferSize(&numOfBufferFrames);
EXIT_ERROR(hr)
hr = audioClient->GetService(local_IID_IAudioCaptureClient, (void**)&audioCaptureClient);
EXIT_ERROR(hr)
actualDuration = (REFERENCE_TIME)((double)REFTIMES_PER_SEC * numOfBufferFrames / waveFormat->nSamplesPerSec);
hr = device->Activate(__uuidof(IAudioEndpointVolume),
CLSCTX_ALL, NULL, (void**)&endpointvol);
hr = endpointvol->GetMasterVolumeLevelScalar(&vol);
std::cout << "master volume in float: " << vol;
hr = endpointvol->SetMasterVolumeLevelScalar(1.0, NULL);
EXIT_ERROR(hr)
hr = endpointvol->GetMasterVolumeLevelScalar(&vol);
std::cout << endl << "master volume in float(after udpate): " << vol;
hr = audioClient->Start();
EXIT_ERROR(hr)
while (done == FALSE) {
Sleep((DWORD)actualDuration / 100000 / 2);
UINT32 packetLength;
PBYTE packet;
UINT32 numFramesAvailable;
DWORD flags;
hr = audioCaptureClient->GetNextPacketSize(&packetLength);
EXIT_ERROR(hr)
while (packetLength != 0) {
if (framesRead > 10 * samplesPerSec) {
done = TRUE;
}
hr = audioCaptureClient->GetBuffer(&packet, &numFramesAvailable, &flags, NULL, NULL);
EXIT_ERROR(hr)
writeData(packet, numFramesAvailable * channels * bytesPerSample);
framesRead += numFramesAvailable;
samples.push_samples((const float*)packet, numFramesAvailable);
hr = audioCaptureClient->ReleaseBuffer(numFramesAvailable);
EXIT_ERROR(hr)
hr = audioCaptureClient->GetNextPacketSize(&packetLength);
EXIT_ERROR(hr)
}
}
endData(framesRead);
EXIT:
RELEASE(properties)
RELEASE(device)
RELEASE(pEnumerator)
RELEASE(audioClient)
RELEASE(audioCaptureClient)
CoTaskMemFree(waveFormat);
Windows::Foundation::Uninitialize();
}
}audiOBJ;
int main() {
struct whisper_context* ctx = whisper_init_from_file("whisper_medium/ggml-base.en.bin");
if (whisper_full(ctx, whisper_full_default_params(WHISPER_SAMPLING_BEAM_SEARCH), samples.buf.data(), samples.buf.size()) != 0) {
fprintf(stderr, "Full Whisper processing failed\n");
return 7;
}
const int n_segments = whisper_full_n_segments(ctx);
for (int i = 0; i < n_segments; ++i) {
const char* text = whisper_full_get_segment_text(ctx, i);
printf("%s", text);
}
whisper_free(ctx);
return 0;
}