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
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
|
/*
* Copyright (c) 2014 The WebRTC project authors. All Rights Reserved.
*
* Use of this source code is governed by a BSD-style license
* that can be found in the LICENSE file in the root of the source
* tree. An additional intellectual property rights grant can be found
* in the file PATENTS. All contributing project authors may
* be found in the AUTHORS file in the root of the source tree.
*/
#include "modules/audio_processing/audio_processing_impl.h"
#include <array>
#include <memory>
#include "api/make_ref_counted.h"
#include "api/scoped_refptr.h"
#include "modules/audio_processing/include/audio_processing.h"
#include "modules/audio_processing/optionally_built_submodule_creators.h"
#include "modules/audio_processing/test/audio_processing_builder_for_testing.h"
#include "modules/audio_processing/test/echo_canceller_test_tools.h"
#include "modules/audio_processing/test/echo_control_mock.h"
#include "modules/audio_processing/test/test_utils.h"
#include "rtc_base/checks.h"
#include "rtc_base/random.h"
#include "test/field_trial.h"
#include "test/gmock.h"
#include "test/gtest.h"
namespace webrtc {
namespace {
using ::testing::Invoke;
using ::testing::NotNull;
class MockInitialize : public AudioProcessingImpl {
public:
MockInitialize() : AudioProcessingImpl() {}
MOCK_METHOD(void, InitializeLocked, (), (override));
void RealInitializeLocked() {
AssertLockedForTest();
AudioProcessingImpl::InitializeLocked();
}
MOCK_METHOD(void, AddRef, (), (const, override));
MOCK_METHOD(rtc::RefCountReleaseStatus, Release, (), (const, override));
};
// Creates MockEchoControl instances and provides a raw pointer access to
// the next created one. The raw pointer is meant to be used with gmock.
// Returning a pointer of the next created MockEchoControl instance is necessary
// for the following reasons: (i) gmock expectations must be set before any call
// occurs, (ii) APM is initialized the first time that
// AudioProcessingImpl::ProcessStream() is called and the initialization leads
// to the creation of a new EchoControl object.
class MockEchoControlFactory : public EchoControlFactory {
public:
MockEchoControlFactory() : next_mock_(std::make_unique<MockEchoControl>()) {}
// Returns a pointer to the next MockEchoControl that this factory creates.
MockEchoControl* GetNext() const { return next_mock_.get(); }
std::unique_ptr<EchoControl> Create(int sample_rate_hz,
int num_render_channels,
int num_capture_channels) override {
std::unique_ptr<EchoControl> mock = std::move(next_mock_);
next_mock_ = std::make_unique<MockEchoControl>();
return mock;
}
private:
std::unique_ptr<MockEchoControl> next_mock_;
};
// Mocks EchoDetector and records the first samples of the last analyzed render
// stream frame. Used to check what data is read by an EchoDetector
// implementation injected into an APM.
class TestEchoDetector : public EchoDetector {
public:
TestEchoDetector()
: analyze_render_audio_called_(false),
last_render_audio_first_sample_(0.f) {}
~TestEchoDetector() override = default;
void AnalyzeRenderAudio(rtc::ArrayView<const float> render_audio) override {
last_render_audio_first_sample_ = render_audio[0];
analyze_render_audio_called_ = true;
}
void AnalyzeCaptureAudio(rtc::ArrayView<const float> capture_audio) override {
}
void Initialize(int capture_sample_rate_hz,
int num_capture_channels,
int render_sample_rate_hz,
int num_render_channels) override {}
EchoDetector::Metrics GetMetrics() const override { return {}; }
// Returns true if AnalyzeRenderAudio() has been called at least once.
bool analyze_render_audio_called() const {
return analyze_render_audio_called_;
}
// Returns the first sample of the last analyzed render frame.
float last_render_audio_first_sample() const {
return last_render_audio_first_sample_;
}
private:
bool analyze_render_audio_called_;
float last_render_audio_first_sample_;
};
// Mocks CustomProcessing and applies ProcessSample() to all the samples.
// Meant to be injected into an APM to modify samples in a known and detectable
// way.
class TestRenderPreProcessor : public CustomProcessing {
public:
TestRenderPreProcessor() = default;
~TestRenderPreProcessor() = default;
void Initialize(int sample_rate_hz, int num_channels) override {}
void Process(AudioBuffer* audio) override {
for (size_t k = 0; k < audio->num_channels(); ++k) {
rtc::ArrayView<float> channel_view(audio->channels()[k],
audio->num_frames());
std::transform(channel_view.begin(), channel_view.end(),
channel_view.begin(), ProcessSample);
}
}
std::string ToString() const override { return "TestRenderPreProcessor"; }
void SetRuntimeSetting(AudioProcessing::RuntimeSetting setting) override {}
// Modifies a sample. This member is used in Process() to modify a frame and
// it is publicly visible to enable tests.
static constexpr float ProcessSample(float x) { return 2.f * x; }
};
} // namespace
TEST(AudioProcessingImplTest, AudioParameterChangeTriggersInit) {
MockInitialize mock;
ON_CALL(mock, InitializeLocked)
.WillByDefault(Invoke(&mock, &MockInitialize::RealInitializeLocked));
EXPECT_CALL(mock, InitializeLocked).Times(1);
mock.Initialize();
constexpr size_t kMaxSampleRateHz = 32000;
constexpr size_t kMaxNumChannels = 2;
std::array<int16_t, kMaxNumChannels * kMaxSampleRateHz / 100> frame;
frame.fill(0);
StreamConfig config(16000, 1);
// Call with the default parameters; there should be an init.
EXPECT_CALL(mock, InitializeLocked).Times(0);
EXPECT_NOERR(mock.ProcessStream(frame.data(), config, config, frame.data()));
EXPECT_NOERR(
mock.ProcessReverseStream(frame.data(), config, config, frame.data()));
// New sample rate. (Only impacts ProcessStream).
config = StreamConfig(32000, 1);
EXPECT_CALL(mock, InitializeLocked).Times(1);
EXPECT_NOERR(mock.ProcessStream(frame.data(), config, config, frame.data()));
// New number of channels.
// TODO(peah): Investigate why this causes 2 inits.
config = StreamConfig(32000, 2);
EXPECT_CALL(mock, InitializeLocked).Times(2);
EXPECT_NOERR(mock.ProcessStream(frame.data(), config, config, frame.data()));
// ProcessStream sets num_channels_ == num_output_channels.
EXPECT_NOERR(
mock.ProcessReverseStream(frame.data(), config, config, frame.data()));
// A new sample rate passed to ProcessReverseStream should cause an init.
config = StreamConfig(16000, 2);
EXPECT_CALL(mock, InitializeLocked).Times(1);
EXPECT_NOERR(
mock.ProcessReverseStream(frame.data(), config, config, frame.data()));
}
TEST(AudioProcessingImplTest, UpdateCapturePreGainRuntimeSetting) {
rtc::scoped_refptr<AudioProcessing> apm =
AudioProcessingBuilderForTesting().Create();
webrtc::AudioProcessing::Config apm_config;
apm_config.pre_amplifier.enabled = true;
apm_config.pre_amplifier.fixed_gain_factor = 1.f;
apm->ApplyConfig(apm_config);
constexpr int kSampleRateHz = 48000;
constexpr int16_t kAudioLevel = 10000;
constexpr size_t kNumChannels = 2;
std::array<int16_t, kNumChannels * kSampleRateHz / 100> frame;
StreamConfig config(kSampleRateHz, kNumChannels);
frame.fill(kAudioLevel);
apm->ProcessStream(frame.data(), config, config, frame.data());
EXPECT_EQ(frame[100], kAudioLevel)
<< "With factor 1, frame shouldn't be modified.";
constexpr float kGainFactor = 2.f;
apm->SetRuntimeSetting(
AudioProcessing::RuntimeSetting::CreateCapturePreGain(kGainFactor));
// Process for two frames to have time to ramp up gain.
for (int i = 0; i < 2; ++i) {
frame.fill(kAudioLevel);
apm->ProcessStream(frame.data(), config, config, frame.data());
}
EXPECT_EQ(frame[100], kGainFactor * kAudioLevel)
<< "Frame should be amplified.";
}
TEST(AudioProcessingImplTest,
LevelAdjustmentUpdateCapturePreGainRuntimeSetting) {
rtc::scoped_refptr<AudioProcessing> apm =
AudioProcessingBuilderForTesting().Create();
webrtc::AudioProcessing::Config apm_config;
apm_config.capture_level_adjustment.enabled = true;
apm_config.capture_level_adjustment.pre_gain_factor = 1.f;
apm->ApplyConfig(apm_config);
constexpr int kSampleRateHz = 48000;
constexpr int16_t kAudioLevel = 10000;
constexpr size_t kNumChannels = 2;
std::array<int16_t, kNumChannels * kSampleRateHz / 100> frame;
StreamConfig config(kSampleRateHz, kNumChannels);
frame.fill(kAudioLevel);
apm->ProcessStream(frame.data(), config, config, frame.data());
EXPECT_EQ(frame[100], kAudioLevel)
<< "With factor 1, frame shouldn't be modified.";
constexpr float kGainFactor = 2.f;
apm->SetRuntimeSetting(
AudioProcessing::RuntimeSetting::CreateCapturePreGain(kGainFactor));
// Process for two frames to have time to ramp up gain.
for (int i = 0; i < 2; ++i) {
frame.fill(kAudioLevel);
apm->ProcessStream(frame.data(), config, config, frame.data());
}
EXPECT_EQ(frame[100], kGainFactor * kAudioLevel)
<< "Frame should be amplified.";
}
TEST(AudioProcessingImplTest,
LevelAdjustmentUpdateCapturePostGainRuntimeSetting) {
rtc::scoped_refptr<AudioProcessing> apm =
AudioProcessingBuilderForTesting().Create();
webrtc::AudioProcessing::Config apm_config;
apm_config.capture_level_adjustment.enabled = true;
apm_config.capture_level_adjustment.post_gain_factor = 1.f;
apm->ApplyConfig(apm_config);
constexpr int kSampleRateHz = 48000;
constexpr int16_t kAudioLevel = 10000;
constexpr size_t kNumChannels = 2;
std::array<int16_t, kNumChannels * kSampleRateHz / 100> frame;
StreamConfig config(kSampleRateHz, kNumChannels);
frame.fill(kAudioLevel);
apm->ProcessStream(frame.data(), config, config, frame.data());
EXPECT_EQ(frame[100], kAudioLevel)
<< "With factor 1, frame shouldn't be modified.";
constexpr float kGainFactor = 2.f;
apm->SetRuntimeSetting(
AudioProcessing::RuntimeSetting::CreateCapturePostGain(kGainFactor));
// Process for two frames to have time to ramp up gain.
for (int i = 0; i < 2; ++i) {
frame.fill(kAudioLevel);
apm->ProcessStream(frame.data(), config, config, frame.data());
}
EXPECT_EQ(frame[100], kGainFactor * kAudioLevel)
<< "Frame should be amplified.";
}
TEST(AudioProcessingImplTest, EchoControllerObservesSetCaptureUsageChange) {
// Tests that the echo controller observes that the capture usage has been
// updated.
auto echo_control_factory = std::make_unique<MockEchoControlFactory>();
const MockEchoControlFactory* echo_control_factory_ptr =
echo_control_factory.get();
rtc::scoped_refptr<AudioProcessing> apm =
AudioProcessingBuilderForTesting()
.SetEchoControlFactory(std::move(echo_control_factory))
.Create();
constexpr int16_t kAudioLevel = 10000;
constexpr int kSampleRateHz = 48000;
constexpr int kNumChannels = 2;
std::array<int16_t, kNumChannels * kSampleRateHz / 100> frame;
StreamConfig config(kSampleRateHz, kNumChannels);
frame.fill(kAudioLevel);
MockEchoControl* echo_control_mock = echo_control_factory_ptr->GetNext();
// Ensure that SetCaptureOutputUsage is not called when no runtime settings
// are passed.
EXPECT_CALL(*echo_control_mock, SetCaptureOutputUsage(testing::_)).Times(0);
apm->ProcessStream(frame.data(), config, config, frame.data());
// Ensure that SetCaptureOutputUsage is called with the right information when
// a runtime setting is passed.
EXPECT_CALL(*echo_control_mock,
SetCaptureOutputUsage(/*capture_output_used=*/false))
.Times(1);
EXPECT_TRUE(apm->PostRuntimeSetting(
AudioProcessing::RuntimeSetting::CreateCaptureOutputUsedSetting(
/*capture_output_used=*/false)));
apm->ProcessStream(frame.data(), config, config, frame.data());
EXPECT_CALL(*echo_control_mock,
SetCaptureOutputUsage(/*capture_output_used=*/true))
.Times(1);
EXPECT_TRUE(apm->PostRuntimeSetting(
AudioProcessing::RuntimeSetting::CreateCaptureOutputUsedSetting(
/*capture_output_used=*/true)));
apm->ProcessStream(frame.data(), config, config, frame.data());
// The number of positions to place items in the queue is equal to the queue
// size minus 1.
constexpr int kNumSlotsInQueue = RuntimeSettingQueueSize();
// Ensure that SetCaptureOutputUsage is called with the right information when
// many runtime settings are passed.
for (int k = 0; k < kNumSlotsInQueue - 1; ++k) {
EXPECT_TRUE(apm->PostRuntimeSetting(
AudioProcessing::RuntimeSetting::CreateCaptureOutputUsedSetting(
/*capture_output_used=*/false)));
}
EXPECT_CALL(*echo_control_mock,
SetCaptureOutputUsage(/*capture_output_used=*/false))
.Times(kNumSlotsInQueue - 1);
apm->ProcessStream(frame.data(), config, config, frame.data());
// Ensure that SetCaptureOutputUsage is properly called with the fallback
// value when the runtime settings queue becomes full.
for (int k = 0; k < kNumSlotsInQueue; ++k) {
EXPECT_TRUE(apm->PostRuntimeSetting(
AudioProcessing::RuntimeSetting::CreateCaptureOutputUsedSetting(
/*capture_output_used=*/false)));
}
EXPECT_FALSE(apm->PostRuntimeSetting(
AudioProcessing::RuntimeSetting::CreateCaptureOutputUsedSetting(
/*capture_output_used=*/false)));
EXPECT_FALSE(apm->PostRuntimeSetting(
AudioProcessing::RuntimeSetting::CreateCaptureOutputUsedSetting(
/*capture_output_used=*/false)));
EXPECT_CALL(*echo_control_mock,
SetCaptureOutputUsage(/*capture_output_used=*/false))
.Times(kNumSlotsInQueue);
EXPECT_CALL(*echo_control_mock,
SetCaptureOutputUsage(/*capture_output_used=*/true))
.Times(1);
apm->ProcessStream(frame.data(), config, config, frame.data());
}
TEST(AudioProcessingImplTest,
EchoControllerObservesPreAmplifierEchoPathGainChange) {
// Tests that the echo controller observes an echo path gain change when the
// pre-amplifier submodule changes the gain.
auto echo_control_factory = std::make_unique<MockEchoControlFactory>();
const auto* echo_control_factory_ptr = echo_control_factory.get();
rtc::scoped_refptr<AudioProcessing> apm =
AudioProcessingBuilderForTesting()
.SetEchoControlFactory(std::move(echo_control_factory))
.Create();
// Disable AGC.
webrtc::AudioProcessing::Config apm_config;
apm_config.gain_controller1.enabled = false;
apm_config.gain_controller2.enabled = false;
apm_config.pre_amplifier.enabled = true;
apm_config.pre_amplifier.fixed_gain_factor = 1.f;
apm->ApplyConfig(apm_config);
constexpr int16_t kAudioLevel = 10000;
constexpr size_t kSampleRateHz = 48000;
constexpr size_t kNumChannels = 2;
std::array<int16_t, kNumChannels * kSampleRateHz / 100> frame;
StreamConfig config(kSampleRateHz, kNumChannels);
frame.fill(kAudioLevel);
MockEchoControl* echo_control_mock = echo_control_factory_ptr->GetNext();
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock,
ProcessCapture(NotNull(), testing::_, /*echo_path_change=*/false))
.Times(1);
apm->ProcessStream(frame.data(), config, config, frame.data());
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock,
ProcessCapture(NotNull(), testing::_, /*echo_path_change=*/true))
.Times(1);
apm->SetRuntimeSetting(
AudioProcessing::RuntimeSetting::CreateCapturePreGain(2.f));
apm->ProcessStream(frame.data(), config, config, frame.data());
}
TEST(AudioProcessingImplTest,
EchoControllerObservesLevelAdjustmentPreGainEchoPathGainChange) {
// Tests that the echo controller observes an echo path gain change when the
// pre-amplifier submodule changes the gain.
auto echo_control_factory = std::make_unique<MockEchoControlFactory>();
const auto* echo_control_factory_ptr = echo_control_factory.get();
rtc::scoped_refptr<AudioProcessing> apm =
AudioProcessingBuilderForTesting()
.SetEchoControlFactory(std::move(echo_control_factory))
.Create();
// Disable AGC.
webrtc::AudioProcessing::Config apm_config;
apm_config.gain_controller1.enabled = false;
apm_config.gain_controller2.enabled = false;
apm_config.capture_level_adjustment.enabled = true;
apm_config.capture_level_adjustment.pre_gain_factor = 1.f;
apm->ApplyConfig(apm_config);
constexpr int16_t kAudioLevel = 10000;
constexpr size_t kSampleRateHz = 48000;
constexpr size_t kNumChannels = 2;
std::array<int16_t, kNumChannels * kSampleRateHz / 100> frame;
StreamConfig config(kSampleRateHz, kNumChannels);
frame.fill(kAudioLevel);
MockEchoControl* echo_control_mock = echo_control_factory_ptr->GetNext();
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock,
ProcessCapture(NotNull(), testing::_, /*echo_path_change=*/false))
.Times(1);
apm->ProcessStream(frame.data(), config, config, frame.data());
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock,
ProcessCapture(NotNull(), testing::_, /*echo_path_change=*/true))
.Times(1);
apm->SetRuntimeSetting(
AudioProcessing::RuntimeSetting::CreateCapturePreGain(2.f));
apm->ProcessStream(frame.data(), config, config, frame.data());
}
TEST(AudioProcessingImplTest,
EchoControllerObservesAnalogAgc1EchoPathGainChange) {
// Tests that the echo controller observes an echo path gain change when the
// AGC1 analog adaptive submodule changes the analog gain.
auto echo_control_factory = std::make_unique<MockEchoControlFactory>();
const auto* echo_control_factory_ptr = echo_control_factory.get();
rtc::scoped_refptr<AudioProcessing> apm =
AudioProcessingBuilderForTesting()
.SetEchoControlFactory(std::move(echo_control_factory))
.Create();
webrtc::AudioProcessing::Config apm_config;
// Enable AGC1.
apm_config.gain_controller1.enabled = true;
apm_config.gain_controller1.mode =
AudioProcessing::Config::GainController1::kAdaptiveAnalog;
apm_config.gain_controller2.enabled = false;
apm_config.pre_amplifier.enabled = false;
apm->ApplyConfig(apm_config);
constexpr int16_t kAudioLevel = 1000;
constexpr size_t kSampleRateHz = 48000;
constexpr size_t kNumChannels = 2;
std::array<int16_t, kNumChannels * kSampleRateHz / 100> frame;
StreamConfig stream_config(kSampleRateHz, kNumChannels);
frame.fill(kAudioLevel);
MockEchoControl* echo_control_mock = echo_control_factory_ptr->GetNext();
const int initial_analog_gain = apm->recommended_stream_analog_level();
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock, ProcessCapture(NotNull(), testing::_, false))
.Times(1);
apm->ProcessStream(frame.data(), stream_config, stream_config, frame.data());
// Force an analog gain change if it did not happen.
if (initial_analog_gain == apm->recommended_stream_analog_level()) {
apm->set_stream_analog_level(initial_analog_gain + 1);
}
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock, ProcessCapture(NotNull(), testing::_, true))
.Times(1);
apm->ProcessStream(frame.data(), stream_config, stream_config, frame.data());
}
TEST(AudioProcessingImplTest,
EchoControllerObservesNoDigitalAgc2EchoPathGainChange) {
// Tests that the echo controller doesn't observe an echo path gain change
// when the AGC2 digital submodule changes the digital gain without analog
// gain changes.
auto echo_control_factory = std::make_unique<MockEchoControlFactory>();
const auto* echo_control_factory_ptr = echo_control_factory.get();
rtc::scoped_refptr<AudioProcessing> apm =
AudioProcessingBuilderForTesting()
.SetEchoControlFactory(std::move(echo_control_factory))
.Create();
webrtc::AudioProcessing::Config apm_config;
// Disable AGC1 analog.
apm_config.gain_controller1.enabled = false;
// Enable AGC2 digital.
apm_config.gain_controller2.enabled = true;
apm_config.gain_controller2.adaptive_digital.enabled = true;
apm->ApplyConfig(apm_config);
constexpr int16_t kAudioLevel = 1000;
constexpr size_t kSampleRateHz = 48000;
constexpr size_t kNumChannels = 2;
std::array<int16_t, kNumChannels * kSampleRateHz / 100> frame;
StreamConfig stream_config(kSampleRateHz, kNumChannels);
frame.fill(kAudioLevel);
MockEchoControl* echo_control_mock = echo_control_factory_ptr->GetNext();
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock, ProcessCapture(NotNull(), testing::_,
/*echo_path_change=*/false))
.Times(1);
apm->ProcessStream(frame.data(), stream_config, stream_config, frame.data());
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock, ProcessCapture(NotNull(), testing::_,
/*echo_path_change=*/false))
.Times(1);
apm->ProcessStream(frame.data(), stream_config, stream_config, frame.data());
}
TEST(AudioProcessingImplTest, ProcessWithAgc2InjectedSpeechProbability) {
// Tests that a stream is successfully processed for the field trial
// `WebRTC-Audio-TransientSuppressorVadMode/Enabled-RnnVad/` using
// injected speech probability in AGC2 digital.
webrtc::test::ScopedFieldTrials field_trials(
"WebRTC-Audio-TransientSuppressorVadMode/Enabled-RnnVad/");
rtc::scoped_refptr<AudioProcessing> apm = AudioProcessingBuilder().Create();
ASSERT_EQ(apm->Initialize(), AudioProcessing::kNoError);
webrtc::AudioProcessing::Config apm_config;
// Disable AGC1 analog.
apm_config.gain_controller1.enabled = false;
// Enable AGC2 digital.
apm_config.gain_controller2.enabled = true;
apm_config.gain_controller2.adaptive_digital.enabled = true;
apm->ApplyConfig(apm_config);
constexpr int kSampleRateHz = 48000;
constexpr int kNumChannels = 1;
std::array<float, kSampleRateHz / 100> buffer;
float* channel_pointers[] = {buffer.data()};
StreamConfig stream_config(/*sample_rate_hz=*/kSampleRateHz,
/*num_channels=*/kNumChannels);
Random random_generator(2341U);
constexpr int kFramesToProcess = 10;
for (int i = 0; i < kFramesToProcess; ++i) {
RandomizeSampleVector(&random_generator, buffer);
ASSERT_EQ(apm->ProcessStream(channel_pointers, stream_config, stream_config,
channel_pointers),
kNoErr);
}
}
TEST(AudioProcessingImplTest, EchoControllerObservesPlayoutVolumeChange) {
// Tests that the echo controller observes an echo path gain change when a
// playout volume change is reported.
auto echo_control_factory = std::make_unique<MockEchoControlFactory>();
const auto* echo_control_factory_ptr = echo_control_factory.get();
rtc::scoped_refptr<AudioProcessing> apm =
AudioProcessingBuilderForTesting()
.SetEchoControlFactory(std::move(echo_control_factory))
.Create();
// Disable AGC.
webrtc::AudioProcessing::Config apm_config;
apm_config.gain_controller1.enabled = false;
apm_config.gain_controller2.enabled = false;
apm->ApplyConfig(apm_config);
constexpr int16_t kAudioLevel = 10000;
constexpr size_t kSampleRateHz = 48000;
constexpr size_t kNumChannels = 2;
std::array<int16_t, kNumChannels * kSampleRateHz / 100> frame;
StreamConfig stream_config(kSampleRateHz, kNumChannels);
frame.fill(kAudioLevel);
MockEchoControl* echo_control_mock = echo_control_factory_ptr->GetNext();
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock,
ProcessCapture(NotNull(), testing::_, /*echo_path_change=*/false))
.Times(1);
apm->ProcessStream(frame.data(), stream_config, stream_config, frame.data());
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock,
ProcessCapture(NotNull(), testing::_, /*echo_path_change=*/false))
.Times(1);
apm->SetRuntimeSetting(
AudioProcessing::RuntimeSetting::CreatePlayoutVolumeChange(50));
apm->ProcessStream(frame.data(), stream_config, stream_config, frame.data());
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock,
ProcessCapture(NotNull(), testing::_, /*echo_path_change=*/false))
.Times(1);
apm->SetRuntimeSetting(
AudioProcessing::RuntimeSetting::CreatePlayoutVolumeChange(50));
apm->ProcessStream(frame.data(), stream_config, stream_config, frame.data());
EXPECT_CALL(*echo_control_mock, AnalyzeCapture(testing::_)).Times(1);
EXPECT_CALL(*echo_control_mock,
ProcessCapture(NotNull(), testing::_, /*echo_path_change=*/true))
.Times(1);
apm->SetRuntimeSetting(
AudioProcessing::RuntimeSetting::CreatePlayoutVolumeChange(100));
apm->ProcessStream(frame.data(), stream_config, stream_config, frame.data());
}
TEST(AudioProcessingImplTest, RenderPreProcessorBeforeEchoDetector) {
// Make sure that signal changes caused by a render pre-processing sub-module
// take place before any echo detector analysis.
auto test_echo_detector = rtc::make_ref_counted<TestEchoDetector>();
std::unique_ptr<CustomProcessing> test_render_pre_processor(
new TestRenderPreProcessor());
// Create APM injecting the test echo detector and render pre-processor.
rtc::scoped_refptr<AudioProcessing> apm =
AudioProcessingBuilderForTesting()
.SetEchoDetector(test_echo_detector)
.SetRenderPreProcessing(std::move(test_render_pre_processor))
.Create();
webrtc::AudioProcessing::Config apm_config;
apm_config.pre_amplifier.enabled = true;
apm->ApplyConfig(apm_config);
constexpr int16_t kAudioLevel = 1000;
constexpr int kSampleRateHz = 16000;
constexpr size_t kNumChannels = 1;
// Explicitly initialize APM to ensure no render frames are discarded.
const ProcessingConfig processing_config = {{
{kSampleRateHz, kNumChannels},
{kSampleRateHz, kNumChannels},
{kSampleRateHz, kNumChannels},
{kSampleRateHz, kNumChannels},
}};
apm->Initialize(processing_config);
std::array<int16_t, kNumChannels * kSampleRateHz / 100> frame;
StreamConfig stream_config(kSampleRateHz, kNumChannels);
constexpr float kAudioLevelFloat = static_cast<float>(kAudioLevel);
constexpr float kExpectedPreprocessedAudioLevel =
TestRenderPreProcessor::ProcessSample(kAudioLevelFloat);
ASSERT_NE(kAudioLevelFloat, kExpectedPreprocessedAudioLevel);
// Analyze a render stream frame.
frame.fill(kAudioLevel);
ASSERT_EQ(AudioProcessing::Error::kNoError,
apm->ProcessReverseStream(frame.data(), stream_config,
stream_config, frame.data()));
// Trigger a call to in EchoDetector::AnalyzeRenderAudio() via
// ProcessStream().
frame.fill(kAudioLevel);
ASSERT_EQ(AudioProcessing::Error::kNoError,
apm->ProcessStream(frame.data(), stream_config, stream_config,
frame.data()));
// Regardless of how the call to in EchoDetector::AnalyzeRenderAudio() is
// triggered, the line below checks that the call has occurred. If not, the
// APM implementation may have changed and this test might need to be adapted.
ASSERT_TRUE(test_echo_detector->analyze_render_audio_called());
// Check that the data read in EchoDetector::AnalyzeRenderAudio() is that
// produced by the render pre-processor.
EXPECT_EQ(kExpectedPreprocessedAudioLevel,
test_echo_detector->last_render_audio_first_sample());
}
// Disabling build-optional submodules and trying to enable them via the APM
// config should be bit-exact with running APM with said submodules disabled.
// This mainly tests that SetCreateOptionalSubmodulesForTesting has an effect.
TEST(ApmWithSubmodulesExcludedTest, BitexactWithDisabledModules) {
auto apm = rtc::make_ref_counted<AudioProcessingImpl>();
ASSERT_EQ(apm->Initialize(), AudioProcessing::kNoError);
ApmSubmoduleCreationOverrides overrides;
overrides.transient_suppression = true;
apm->OverrideSubmoduleCreationForTesting(overrides);
AudioProcessing::Config apm_config = apm->GetConfig();
apm_config.transient_suppression.enabled = true;
apm->ApplyConfig(apm_config);
rtc::scoped_refptr<AudioProcessing> apm_reference =
AudioProcessingBuilder().Create();
apm_config = apm_reference->GetConfig();
apm_config.transient_suppression.enabled = false;
apm_reference->ApplyConfig(apm_config);
constexpr int kSampleRateHz = 16000;
constexpr int kNumChannels = 1;
std::array<float, kSampleRateHz / 100> buffer;
std::array<float, kSampleRateHz / 100> buffer_reference;
float* channel_pointers[] = {buffer.data()};
float* channel_pointers_reference[] = {buffer_reference.data()};
StreamConfig stream_config(/*sample_rate_hz=*/kSampleRateHz,
/*num_channels=*/kNumChannels);
Random random_generator(2341U);
constexpr int kFramesToProcessPerConfiguration = 10;
for (int i = 0; i < kFramesToProcessPerConfiguration; ++i) {
RandomizeSampleVector(&random_generator, buffer);
std::copy(buffer.begin(), buffer.end(), buffer_reference.begin());
ASSERT_EQ(apm->ProcessStream(channel_pointers, stream_config, stream_config,
channel_pointers),
kNoErr);
ASSERT_EQ(
apm_reference->ProcessStream(channel_pointers_reference, stream_config,
stream_config, channel_pointers_reference),
kNoErr);
for (int j = 0; j < kSampleRateHz / 100; ++j) {
EXPECT_EQ(buffer[j], buffer_reference[j]);
}
}
}
// Disable transient suppressor creation and run APM in ways that should trigger
// calls to the transient suppressor API.
TEST(ApmWithSubmodulesExcludedTest, ReinitializeTransientSuppressor) {
auto apm = rtc::make_ref_counted<AudioProcessingImpl>();
ASSERT_EQ(apm->Initialize(), kNoErr);
ApmSubmoduleCreationOverrides overrides;
overrides.transient_suppression = true;
apm->OverrideSubmoduleCreationForTesting(overrides);
AudioProcessing::Config config = apm->GetConfig();
config.transient_suppression.enabled = true;
apm->ApplyConfig(config);
// 960 samples per frame: 10 ms of <= 48 kHz audio with <= 2 channels.
float buffer[960];
float* channel_pointers[] = {&buffer[0], &buffer[480]};
Random random_generator(2341U);
constexpr int kFramesToProcessPerConfiguration = 3;
StreamConfig initial_stream_config(/*sample_rate_hz=*/16000,
/*num_channels=*/1);
for (int i = 0; i < kFramesToProcessPerConfiguration; ++i) {
RandomizeSampleVector(&random_generator, buffer);
EXPECT_EQ(apm->ProcessStream(channel_pointers, initial_stream_config,
initial_stream_config, channel_pointers),
kNoErr);
}
StreamConfig stereo_stream_config(/*sample_rate_hz=*/16000,
/*num_channels=*/2);
for (int i = 0; i < kFramesToProcessPerConfiguration; ++i) {
RandomizeSampleVector(&random_generator, buffer);
EXPECT_EQ(apm->ProcessStream(channel_pointers, stereo_stream_config,
stereo_stream_config, channel_pointers),
kNoErr);
}
StreamConfig high_sample_rate_stream_config(/*sample_rate_hz=*/48000,
/*num_channels=*/2);
for (int i = 0; i < kFramesToProcessPerConfiguration; ++i) {
RandomizeSampleVector(&random_generator, buffer);
EXPECT_EQ(
apm->ProcessStream(channel_pointers, high_sample_rate_stream_config,
high_sample_rate_stream_config, channel_pointers),
kNoErr);
}
}
// Disable transient suppressor creation and run APM in ways that should trigger
// calls to the transient suppressor API.
TEST(ApmWithSubmodulesExcludedTest, ToggleTransientSuppressor) {
auto apm = rtc::make_ref_counted<AudioProcessingImpl>();
ASSERT_EQ(apm->Initialize(), AudioProcessing::kNoError);
ApmSubmoduleCreationOverrides overrides;
overrides.transient_suppression = true;
apm->OverrideSubmoduleCreationForTesting(overrides);
// 960 samples per frame: 10 ms of <= 48 kHz audio with <= 2 channels.
float buffer[960];
float* channel_pointers[] = {&buffer[0], &buffer[480]};
Random random_generator(2341U);
constexpr int kFramesToProcessPerConfiguration = 3;
StreamConfig stream_config(/*sample_rate_hz=*/16000,
/*num_channels=*/1);
AudioProcessing::Config config = apm->GetConfig();
config.transient_suppression.enabled = true;
apm->ApplyConfig(config);
for (int i = 0; i < kFramesToProcessPerConfiguration; ++i) {
RandomizeSampleVector(&random_generator, buffer);
EXPECT_EQ(apm->ProcessStream(channel_pointers, stream_config, stream_config,
channel_pointers),
kNoErr);
}
config = apm->GetConfig();
config.transient_suppression.enabled = false;
apm->ApplyConfig(config);
for (int i = 0; i < kFramesToProcessPerConfiguration; ++i) {
RandomizeSampleVector(&random_generator, buffer);
EXPECT_EQ(apm->ProcessStream(channel_pointers, stream_config, stream_config,
channel_pointers),
kNoErr);
}
config = apm->GetConfig();
config.transient_suppression.enabled = true;
apm->ApplyConfig(config);
for (int i = 0; i < kFramesToProcessPerConfiguration; ++i) {
RandomizeSampleVector(&random_generator, buffer);
EXPECT_EQ(apm->ProcessStream(channel_pointers, stream_config, stream_config,
channel_pointers),
kNoErr);
}
}
} // namespace webrtc
|