summaryrefslogtreecommitdiffstats
path: root/remoting/host/audio_capturer_win.cc
blob: 8c901137945dd58a475a9faca8dfcf36b92ed237 (plain)
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
// Copyright (c) 2012 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

#include "remoting/host/audio_capturer_win.h"

#include <avrt.h>
#include <mmreg.h>
#include <mmsystem.h>
#include <stdint.h>
#include <stdlib.h>
#include <windows.h>

#include <algorithm>
#include <utility>

#include "base/logging.h"

namespace {
const int kChannels = 2;
const int kBytesPerSample = 2;
const int kBitsPerSample = kBytesPerSample * 8;
// Conversion factor from 100ns to 1ms.
const int k100nsPerMillisecond = 10000;

// Tolerance for catching packets of silence. If all samples have absolute
// value less than this threshold, the packet will be counted as a packet of
// silence. A value of 2 was chosen, because Windows can give samples of 1 and
// -1, even when no audio is playing.
const int kSilenceThreshold = 2;

// Lower bound for timer intervals, in milliseconds.
const int kMinTimerInterval = 30;

// Upper bound for the timer precision error, in milliseconds.
// Timers are supposed to be accurate to 20ms, so we use 30ms to be safe.
const int kMaxExpectedTimerLag = 30;
}  // namespace

namespace remoting {

AudioCapturerWin::AudioCapturerWin()
    : sampling_rate_(AudioPacket::SAMPLING_RATE_INVALID),
      silence_detector_(kSilenceThreshold),
      last_capture_error_(S_OK) {
    thread_checker_.DetachFromThread();
}

AudioCapturerWin::~AudioCapturerWin() {
  DCHECK(thread_checker_.CalledOnValidThread());
}

bool AudioCapturerWin::Start(const PacketCapturedCallback& callback) {
  DCHECK(!audio_capture_client_.get());
  DCHECK(!audio_client_.get());
  DCHECK(!mm_device_.get());
  DCHECK(static_cast<PWAVEFORMATEX>(wave_format_ex_) == nullptr);
  DCHECK(thread_checker_.CalledOnValidThread());

  callback_ = callback;

  // Initialize the capture timer.
  capture_timer_.reset(new base::RepeatingTimer());

  HRESULT hr = S_OK;

  base::win::ScopedComPtr<IMMDeviceEnumerator> mm_device_enumerator;
  hr = mm_device_enumerator.CreateInstance(__uuidof(MMDeviceEnumerator));
  if (FAILED(hr)) {
    LOG(ERROR) << "Failed to create IMMDeviceEnumerator. Error " << hr;
    return false;
  }

  // Get the audio endpoint.
  hr = mm_device_enumerator->GetDefaultAudioEndpoint(eRender,
                                                     eConsole,
                                                     mm_device_.Receive());
  if (FAILED(hr)) {
    LOG(ERROR) << "Failed to get IMMDevice. Error " << hr;
    return false;
  }

  // Get an audio client.
  hr = mm_device_->Activate(__uuidof(IAudioClient),
                            CLSCTX_ALL,
                            nullptr,
                            audio_client_.ReceiveVoid());
  if (FAILED(hr)) {
    LOG(ERROR) << "Failed to get an IAudioClient. Error " << hr;
    return false;
  }

  REFERENCE_TIME device_period;
  hr = audio_client_->GetDevicePeriod(&device_period, nullptr);
  if (FAILED(hr)) {
    LOG(ERROR) << "IAudioClient::GetDevicePeriod failed. Error " << hr;
    return false;
  }
  // We round up, if |device_period| / |k100nsPerMillisecond|
  // is not a whole number.
  int device_period_in_milliseconds =
      1 + ((device_period - 1) / k100nsPerMillisecond);
  audio_device_period_ = base::TimeDelta::FromMilliseconds(
      std::max(device_period_in_milliseconds, kMinTimerInterval));

  // Get the wave format.
  hr = audio_client_->GetMixFormat(&wave_format_ex_);
  if (FAILED(hr)) {
    LOG(ERROR) << "Failed to get WAVEFORMATEX. Error " << hr;
    return false;
  }

  // Set the wave format
  switch (wave_format_ex_->wFormatTag) {
    case WAVE_FORMAT_IEEE_FLOAT:
      // Intentional fall-through.
    case WAVE_FORMAT_PCM:
      if (!AudioCapturer::IsValidSampleRate(wave_format_ex_->nSamplesPerSec)) {
        LOG(ERROR) << "Host sampling rate is neither 44.1 kHz nor 48 kHz.";
        return false;
      }
      sampling_rate_ = static_cast<AudioPacket::SamplingRate>(
          wave_format_ex_->nSamplesPerSec);

      wave_format_ex_->wFormatTag = WAVE_FORMAT_PCM;
      wave_format_ex_->nChannels = kChannels;
      wave_format_ex_->wBitsPerSample = kBitsPerSample;
      wave_format_ex_->nBlockAlign = kChannels * kBytesPerSample;
      wave_format_ex_->nAvgBytesPerSec =
          sampling_rate_ * kChannels * kBytesPerSample;
      break;
    case WAVE_FORMAT_EXTENSIBLE: {
      PWAVEFORMATEXTENSIBLE wave_format_extensible =
          reinterpret_cast<WAVEFORMATEXTENSIBLE*>(
          static_cast<WAVEFORMATEX*>(wave_format_ex_));
      if (IsEqualGUID(KSDATAFORMAT_SUBTYPE_IEEE_FLOAT,
                      wave_format_extensible->SubFormat)) {
        if (!AudioCapturer::IsValidSampleRate(
                wave_format_extensible->Format.nSamplesPerSec)) {
          LOG(ERROR) << "Host sampling rate is neither 44.1 kHz nor 48 kHz.";
          return false;
        }
        sampling_rate_ = static_cast<AudioPacket::SamplingRate>(
            wave_format_extensible->Format.nSamplesPerSec);

        wave_format_extensible->SubFormat = KSDATAFORMAT_SUBTYPE_PCM;
        wave_format_extensible->Samples.wValidBitsPerSample = kBitsPerSample;

        wave_format_extensible->Format.nChannels = kChannels;
        wave_format_extensible->Format.nSamplesPerSec = sampling_rate_;
        wave_format_extensible->Format.wBitsPerSample = kBitsPerSample;
        wave_format_extensible->Format.nBlockAlign =
            kChannels * kBytesPerSample;
        wave_format_extensible->Format.nAvgBytesPerSec =
            sampling_rate_ * kChannels * kBytesPerSample;
      } else {
        LOG(ERROR) << "Failed to force 16-bit samples";
        return false;
      }
      break;
    }
    default:
      LOG(ERROR) << "Failed to force 16-bit PCM";
      return false;
  }

  // Initialize the IAudioClient.
  hr = audio_client_->Initialize(
      AUDCLNT_SHAREMODE_SHARED,
      AUDCLNT_STREAMFLAGS_LOOPBACK,
      (kMaxExpectedTimerLag + audio_device_period_.InMilliseconds()) *
      k100nsPerMillisecond,
      0,
      wave_format_ex_,
      nullptr);
  if (FAILED(hr)) {
    LOG(ERROR) << "Failed to initialize IAudioClient. Error " << hr;
    return false;
  }

  // Get an IAudioCaptureClient.
  hr = audio_client_->GetService(__uuidof(IAudioCaptureClient),
                                 audio_capture_client_.ReceiveVoid());
  if (FAILED(hr)) {
    LOG(ERROR) << "Failed to get an IAudioCaptureClient. Error " << hr;
    return false;
  }

  // Start the IAudioClient.
  hr = audio_client_->Start();
  if (FAILED(hr)) {
    LOG(ERROR) << "Failed to start IAudioClient. Error " << hr;
    return false;
  }

  silence_detector_.Reset(sampling_rate_, kChannels);

  // Start capturing.
  capture_timer_->Start(FROM_HERE,
                        audio_device_period_,
                        this,
                        &AudioCapturerWin::DoCapture);
  return true;
}

void AudioCapturerWin::DoCapture() {
  DCHECK(AudioCapturer::IsValidSampleRate(sampling_rate_));
  DCHECK(thread_checker_.CalledOnValidThread());

  // Fetch all packets from the audio capture endpoint buffer.
  HRESULT hr = S_OK;
  while (true) {
    UINT32 next_packet_size;
    HRESULT hr = audio_capture_client_->GetNextPacketSize(&next_packet_size);
    if (FAILED(hr))
      break;

    if (next_packet_size <= 0) {
      return;
    }

    BYTE* data;
    UINT32 frames;
    DWORD flags;
    hr = audio_capture_client_->GetBuffer(&data, &frames, &flags, nullptr,
                                          nullptr);
    if (FAILED(hr))
      break;

    if ((flags & AUDCLNT_BUFFERFLAGS_SILENT) == 0 &&
        !silence_detector_.IsSilence(reinterpret_cast<const int16_t*>(data),
                                     frames * kChannels)) {
      scoped_ptr<AudioPacket> packet(new AudioPacket());
      packet->add_data(data, frames * wave_format_ex_->nBlockAlign);
      packet->set_encoding(AudioPacket::ENCODING_RAW);
      packet->set_sampling_rate(sampling_rate_);
      packet->set_bytes_per_sample(AudioPacket::BYTES_PER_SAMPLE_2);
      packet->set_channels(AudioPacket::CHANNELS_STEREO);

      callback_.Run(std::move(packet));
    }

    hr = audio_capture_client_->ReleaseBuffer(frames);
    if (FAILED(hr))
      break;
  }

  // There is nothing to capture if the audio endpoint device has been unplugged
  // or disabled.
  if (hr == AUDCLNT_E_DEVICE_INVALIDATED)
    return;

  // Avoid reporting the same error multiple times.
  if (FAILED(hr) && hr != last_capture_error_) {
    last_capture_error_ = hr;
    LOG(ERROR) << "Failed to capture an audio packet: 0x"
               << std::hex << hr << std::dec << ".";
  }
}

bool AudioCapturer::IsSupported() {
  return true;
}

scoped_ptr<AudioCapturer> AudioCapturer::Create() {
  return make_scoped_ptr(new AudioCapturerWin());
}

}  // namespace remoting