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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056
|
/*
* Copyright (c) 2012 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 "webrtc/modules/audio_processing/audio_processing_impl.h"
#include <assert.h>
#include "webrtc/base/platform_file.h"
#include "webrtc/common_audio/include/audio_util.h"
#include "webrtc/common_audio/signal_processing/include/signal_processing_library.h"
#include "webrtc/modules/audio_processing/agc/agc_manager_direct.h"
#include "webrtc/modules/audio_processing/audio_buffer.h"
#include "webrtc/modules/audio_processing/beamformer/beamformer.h"
#include "webrtc/modules/audio_processing/channel_buffer.h"
#include "webrtc/modules/audio_processing/common.h"
#include "webrtc/modules/audio_processing/echo_cancellation_impl.h"
#include "webrtc/modules/audio_processing/echo_control_mobile_impl.h"
#include "webrtc/modules/audio_processing/gain_control_impl.h"
#include "webrtc/modules/audio_processing/high_pass_filter_impl.h"
#include "webrtc/modules/audio_processing/level_estimator_impl.h"
#include "webrtc/modules/audio_processing/noise_suppression_impl.h"
#include "webrtc/modules/audio_processing/processing_component.h"
#include "webrtc/modules/audio_processing/transient/transient_suppressor.h"
#include "webrtc/modules/audio_processing/voice_detection_impl.h"
#include "webrtc/modules/interface/module_common_types.h"
#include "webrtc/system_wrappers/interface/compile_assert.h"
#include "webrtc/system_wrappers/interface/critical_section_wrapper.h"
#include "webrtc/system_wrappers/interface/file_wrapper.h"
#include "webrtc/system_wrappers/interface/logging.h"
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
// Files generated at build-time by the protobuf compiler.
#ifdef WEBRTC_ANDROID_PLATFORM_BUILD
#include "external/webrtc/webrtc/modules/audio_processing/debug.pb.h"
#else
#include "webrtc/audio_processing/debug.pb.h"
#endif
#endif // WEBRTC_AUDIOPROC_DEBUG_DUMP
#define RETURN_ON_ERR(expr) \
do { \
int err = expr; \
if (err != kNoError) { \
return err; \
} \
} while (0)
namespace webrtc {
// Throughout webrtc, it's assumed that success is represented by zero.
COMPILE_ASSERT(AudioProcessing::kNoError == 0, no_error_must_be_zero);
// This class has two main functionalities:
//
// 1) It is returned instead of the real GainControl after the new AGC has been
// enabled in order to prevent an outside user from overriding compression
// settings. It doesn't do anything in its implementation, except for
// delegating the const methods and Enable calls to the real GainControl, so
// AGC can still be disabled.
//
// 2) It is injected into AgcManagerDirect and implements volume callbacks for
// getting and setting the volume level. It just caches this value to be used
// in VoiceEngine later.
class GainControlForNewAgc : public GainControl, public VolumeCallbacks {
public:
explicit GainControlForNewAgc(GainControlImpl* gain_control)
: real_gain_control_(gain_control),
volume_(0) {
}
// GainControl implementation.
virtual int Enable(bool enable) OVERRIDE {
return real_gain_control_->Enable(enable);
}
virtual bool is_enabled() const OVERRIDE {
return real_gain_control_->is_enabled();
}
virtual int set_stream_analog_level(int level) OVERRIDE {
volume_ = level;
return AudioProcessing::kNoError;
}
virtual int stream_analog_level() OVERRIDE {
return volume_;
}
virtual int set_mode(Mode mode) OVERRIDE { return AudioProcessing::kNoError; }
virtual Mode mode() const OVERRIDE { return GainControl::kAdaptiveAnalog; }
virtual int set_target_level_dbfs(int level) OVERRIDE {
return AudioProcessing::kNoError;
}
virtual int target_level_dbfs() const OVERRIDE {
return real_gain_control_->target_level_dbfs();
}
virtual int set_compression_gain_db(int gain) OVERRIDE {
return AudioProcessing::kNoError;
}
virtual int compression_gain_db() const OVERRIDE {
return real_gain_control_->compression_gain_db();
}
virtual int enable_limiter(bool enable) OVERRIDE {
return AudioProcessing::kNoError;
}
virtual bool is_limiter_enabled() const OVERRIDE {
return real_gain_control_->is_limiter_enabled();
}
virtual int set_analog_level_limits(int minimum,
int maximum) OVERRIDE {
return AudioProcessing::kNoError;
}
virtual int analog_level_minimum() const OVERRIDE {
return real_gain_control_->analog_level_minimum();
}
virtual int analog_level_maximum() const OVERRIDE {
return real_gain_control_->analog_level_maximum();
}
virtual bool stream_is_saturated() const OVERRIDE {
return real_gain_control_->stream_is_saturated();
}
// VolumeCallbacks implementation.
virtual void SetMicVolume(int volume) OVERRIDE {
volume_ = volume;
}
virtual int GetMicVolume() OVERRIDE {
return volume_;
}
private:
GainControl* real_gain_control_;
int volume_;
};
AudioProcessing* AudioProcessing::Create(int id) {
return Create();
}
AudioProcessing* AudioProcessing::Create() {
Config config;
return Create(config);
}
AudioProcessing* AudioProcessing::Create(const Config& config) {
AudioProcessingImpl* apm = new AudioProcessingImpl(config);
if (apm->Initialize() != kNoError) {
delete apm;
apm = NULL;
}
return apm;
}
AudioProcessingImpl::AudioProcessingImpl(const Config& config)
: echo_cancellation_(NULL),
echo_control_mobile_(NULL),
gain_control_(NULL),
high_pass_filter_(NULL),
level_estimator_(NULL),
noise_suppression_(NULL),
voice_detection_(NULL),
crit_(CriticalSectionWrapper::CreateCriticalSection()),
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
debug_file_(FileWrapper::Create()),
event_msg_(new audioproc::Event()),
#endif
fwd_in_format_(kSampleRate16kHz, 1),
fwd_proc_format_(kSampleRate16kHz),
fwd_out_format_(kSampleRate16kHz, 1),
rev_in_format_(kSampleRate16kHz, 1),
rev_proc_format_(kSampleRate16kHz, 1),
split_rate_(kSampleRate16kHz),
stream_delay_ms_(0),
delay_offset_ms_(0),
was_stream_delay_set_(false),
output_will_be_muted_(false),
key_pressed_(false),
#if defined(WEBRTC_ANDROID) || defined(WEBRTC_IOS)
use_new_agc_(false),
#else
use_new_agc_(config.Get<ExperimentalAgc>().enabled),
#endif
transient_suppressor_enabled_(config.Get<ExperimentalNs>().enabled),
beamformer_enabled_(config.Get<Beamforming>().enabled),
array_geometry_(config.Get<Beamforming>().array_geometry) {
echo_cancellation_ = new EchoCancellationImpl(this, crit_);
component_list_.push_back(echo_cancellation_);
echo_control_mobile_ = new EchoControlMobileImpl(this, crit_);
component_list_.push_back(echo_control_mobile_);
gain_control_ = new GainControlImpl(this, crit_);
component_list_.push_back(gain_control_);
high_pass_filter_ = new HighPassFilterImpl(this, crit_);
component_list_.push_back(high_pass_filter_);
level_estimator_ = new LevelEstimatorImpl(this, crit_);
component_list_.push_back(level_estimator_);
noise_suppression_ = new NoiseSuppressionImpl(this, crit_);
component_list_.push_back(noise_suppression_);
voice_detection_ = new VoiceDetectionImpl(this, crit_);
component_list_.push_back(voice_detection_);
gain_control_for_new_agc_.reset(new GainControlForNewAgc(gain_control_));
SetExtraOptions(config);
}
AudioProcessingImpl::~AudioProcessingImpl() {
{
CriticalSectionScoped crit_scoped(crit_);
// Depends on gain_control_ and gain_control_for_new_agc_.
agc_manager_.reset();
// Depends on gain_control_.
gain_control_for_new_agc_.reset();
while (!component_list_.empty()) {
ProcessingComponent* component = component_list_.front();
component->Destroy();
delete component;
component_list_.pop_front();
}
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
if (debug_file_->Open()) {
debug_file_->CloseFile();
}
#endif
}
delete crit_;
crit_ = NULL;
}
int AudioProcessingImpl::Initialize() {
CriticalSectionScoped crit_scoped(crit_);
return InitializeLocked();
}
int AudioProcessingImpl::set_sample_rate_hz(int rate) {
CriticalSectionScoped crit_scoped(crit_);
return InitializeLocked(rate,
rate,
rev_in_format_.rate(),
fwd_in_format_.num_channels(),
fwd_out_format_.num_channels(),
rev_in_format_.num_channels());
}
int AudioProcessingImpl::Initialize(int input_sample_rate_hz,
int output_sample_rate_hz,
int reverse_sample_rate_hz,
ChannelLayout input_layout,
ChannelLayout output_layout,
ChannelLayout reverse_layout) {
CriticalSectionScoped crit_scoped(crit_);
return InitializeLocked(input_sample_rate_hz,
output_sample_rate_hz,
reverse_sample_rate_hz,
ChannelsFromLayout(input_layout),
ChannelsFromLayout(output_layout),
ChannelsFromLayout(reverse_layout));
}
int AudioProcessingImpl::InitializeLocked() {
const int fwd_audio_buffer_channels = beamformer_enabled_ ?
fwd_in_format_.num_channels() :
fwd_out_format_.num_channels();
render_audio_.reset(new AudioBuffer(rev_in_format_.samples_per_channel(),
rev_in_format_.num_channels(),
rev_proc_format_.samples_per_channel(),
rev_proc_format_.num_channels(),
rev_proc_format_.samples_per_channel()));
capture_audio_.reset(new AudioBuffer(fwd_in_format_.samples_per_channel(),
fwd_in_format_.num_channels(),
fwd_proc_format_.samples_per_channel(),
fwd_audio_buffer_channels,
fwd_out_format_.samples_per_channel()));
// Initialize all components.
std::list<ProcessingComponent*>::iterator it;
for (it = component_list_.begin(); it != component_list_.end(); ++it) {
int err = (*it)->Initialize();
if (err != kNoError) {
return err;
}
}
int err = InitializeExperimentalAgc();
if (err != kNoError) {
return err;
}
err = InitializeTransient();
if (err != kNoError) {
return err;
}
InitializeBeamformer();
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
if (debug_file_->Open()) {
int err = WriteInitMessage();
if (err != kNoError) {
return err;
}
}
#endif
return kNoError;
}
int AudioProcessingImpl::InitializeLocked(int input_sample_rate_hz,
int output_sample_rate_hz,
int reverse_sample_rate_hz,
int num_input_channels,
int num_output_channels,
int num_reverse_channels) {
if (input_sample_rate_hz <= 0 ||
output_sample_rate_hz <= 0 ||
reverse_sample_rate_hz <= 0) {
return kBadSampleRateError;
}
if (num_output_channels > num_input_channels) {
return kBadNumberChannelsError;
}
// Only mono and stereo supported currently.
if (num_input_channels > 2 || num_input_channels < 1 ||
num_output_channels > 2 || num_output_channels < 1 ||
num_reverse_channels > 2 || num_reverse_channels < 1) {
return kBadNumberChannelsError;
}
fwd_in_format_.set(input_sample_rate_hz, num_input_channels);
fwd_out_format_.set(output_sample_rate_hz, num_output_channels);
rev_in_format_.set(reverse_sample_rate_hz, num_reverse_channels);
// We process at the closest native rate >= min(input rate, output rate)...
int min_proc_rate = std::min(fwd_in_format_.rate(), fwd_out_format_.rate());
int fwd_proc_rate;
if (min_proc_rate > kSampleRate16kHz) {
fwd_proc_rate = kSampleRate32kHz;
} else if (min_proc_rate > kSampleRate8kHz) {
fwd_proc_rate = kSampleRate16kHz;
} else {
fwd_proc_rate = kSampleRate8kHz;
}
// ...with one exception.
if (echo_control_mobile_->is_enabled() && min_proc_rate > kSampleRate16kHz) {
fwd_proc_rate = kSampleRate16kHz;
}
fwd_proc_format_.set(fwd_proc_rate);
// We normally process the reverse stream at 16 kHz. Unless...
int rev_proc_rate = kSampleRate16kHz;
if (fwd_proc_format_.rate() == kSampleRate8kHz) {
// ...the forward stream is at 8 kHz.
rev_proc_rate = kSampleRate8kHz;
} else {
if (rev_in_format_.rate() == kSampleRate32kHz) {
// ...or the input is at 32 kHz, in which case we use the splitting
// filter rather than the resampler.
rev_proc_rate = kSampleRate32kHz;
}
}
// Always downmix the reverse stream to mono for analysis. This has been
// demonstrated to work well for AEC in most practical scenarios.
rev_proc_format_.set(rev_proc_rate, 1);
if (fwd_proc_format_.rate() == kSampleRate32kHz ||
fwd_proc_format_.rate() == kSampleRate48kHz) {
split_rate_ = kSampleRate16kHz;
} else {
split_rate_ = fwd_proc_format_.rate();
}
return InitializeLocked();
}
// Calls InitializeLocked() if any of the audio parameters have changed from
// their current values.
int AudioProcessingImpl::MaybeInitializeLocked(int input_sample_rate_hz,
int output_sample_rate_hz,
int reverse_sample_rate_hz,
int num_input_channels,
int num_output_channels,
int num_reverse_channels) {
if (input_sample_rate_hz == fwd_in_format_.rate() &&
output_sample_rate_hz == fwd_out_format_.rate() &&
reverse_sample_rate_hz == rev_in_format_.rate() &&
num_input_channels == fwd_in_format_.num_channels() &&
num_output_channels == fwd_out_format_.num_channels() &&
num_reverse_channels == rev_in_format_.num_channels()) {
return kNoError;
}
if (beamformer_enabled_ &&
(static_cast<size_t>(num_input_channels) != array_geometry_.size() ||
num_output_channels > 1)) {
return kBadNumberChannelsError;
}
return InitializeLocked(input_sample_rate_hz,
output_sample_rate_hz,
reverse_sample_rate_hz,
num_input_channels,
num_output_channels,
num_reverse_channels);
}
void AudioProcessingImpl::SetExtraOptions(const Config& config) {
CriticalSectionScoped crit_scoped(crit_);
std::list<ProcessingComponent*>::iterator it;
for (it = component_list_.begin(); it != component_list_.end(); ++it)
(*it)->SetExtraOptions(config);
if (transient_suppressor_enabled_ != config.Get<ExperimentalNs>().enabled) {
transient_suppressor_enabled_ = config.Get<ExperimentalNs>().enabled;
InitializeTransient();
}
}
int AudioProcessingImpl::input_sample_rate_hz() const {
CriticalSectionScoped crit_scoped(crit_);
return fwd_in_format_.rate();
}
int AudioProcessingImpl::sample_rate_hz() const {
CriticalSectionScoped crit_scoped(crit_);
return fwd_in_format_.rate();
}
int AudioProcessingImpl::proc_sample_rate_hz() const {
return fwd_proc_format_.rate();
}
int AudioProcessingImpl::proc_split_sample_rate_hz() const {
return split_rate_;
}
int AudioProcessingImpl::num_reverse_channels() const {
return rev_proc_format_.num_channels();
}
int AudioProcessingImpl::num_input_channels() const {
return fwd_in_format_.num_channels();
}
int AudioProcessingImpl::num_output_channels() const {
return fwd_out_format_.num_channels();
}
void AudioProcessingImpl::set_output_will_be_muted(bool muted) {
output_will_be_muted_ = muted;
CriticalSectionScoped lock(crit_);
if (agc_manager_.get()) {
agc_manager_->SetCaptureMuted(output_will_be_muted_);
}
}
bool AudioProcessingImpl::output_will_be_muted() const {
return output_will_be_muted_;
}
int AudioProcessingImpl::ProcessStream(const float* const* src,
int samples_per_channel,
int input_sample_rate_hz,
ChannelLayout input_layout,
int output_sample_rate_hz,
ChannelLayout output_layout,
float* const* dest) {
CriticalSectionScoped crit_scoped(crit_);
if (!src || !dest) {
return kNullPointerError;
}
RETURN_ON_ERR(MaybeInitializeLocked(input_sample_rate_hz,
output_sample_rate_hz,
rev_in_format_.rate(),
ChannelsFromLayout(input_layout),
ChannelsFromLayout(output_layout),
rev_in_format_.num_channels()));
if (samples_per_channel != fwd_in_format_.samples_per_channel()) {
return kBadDataLengthError;
}
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
if (debug_file_->Open()) {
event_msg_->set_type(audioproc::Event::STREAM);
audioproc::Stream* msg = event_msg_->mutable_stream();
const size_t channel_size =
sizeof(float) * fwd_in_format_.samples_per_channel();
for (int i = 0; i < fwd_in_format_.num_channels(); ++i)
msg->add_input_channel(src[i], channel_size);
}
#endif
capture_audio_->CopyFrom(src, samples_per_channel, input_layout);
RETURN_ON_ERR(ProcessStreamLocked());
if (output_copy_needed(is_data_processed())) {
capture_audio_->CopyTo(fwd_out_format_.samples_per_channel(),
output_layout,
dest);
}
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
if (debug_file_->Open()) {
audioproc::Stream* msg = event_msg_->mutable_stream();
const size_t channel_size =
sizeof(float) * fwd_out_format_.samples_per_channel();
for (int i = 0; i < fwd_out_format_.num_channels(); ++i)
msg->add_output_channel(dest[i], channel_size);
RETURN_ON_ERR(WriteMessageToDebugFile());
}
#endif
return kNoError;
}
int AudioProcessingImpl::ProcessStream(AudioFrame* frame) {
CriticalSectionScoped crit_scoped(crit_);
if (!frame) {
return kNullPointerError;
}
// Must be a native rate.
if (frame->sample_rate_hz_ != kSampleRate8kHz &&
frame->sample_rate_hz_ != kSampleRate16kHz &&
frame->sample_rate_hz_ != kSampleRate32kHz &&
frame->sample_rate_hz_ != kSampleRate48kHz) {
return kBadSampleRateError;
}
if (echo_control_mobile_->is_enabled() &&
frame->sample_rate_hz_ > kSampleRate16kHz) {
LOG(LS_ERROR) << "AECM only supports 16 or 8 kHz sample rates";
return kUnsupportedComponentError;
}
// TODO(ajm): The input and output rates and channels are currently
// constrained to be identical in the int16 interface.
RETURN_ON_ERR(MaybeInitializeLocked(frame->sample_rate_hz_,
frame->sample_rate_hz_,
rev_in_format_.rate(),
frame->num_channels_,
frame->num_channels_,
rev_in_format_.num_channels()));
if (frame->samples_per_channel_ != fwd_in_format_.samples_per_channel()) {
return kBadDataLengthError;
}
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
if (debug_file_->Open()) {
event_msg_->set_type(audioproc::Event::STREAM);
audioproc::Stream* msg = event_msg_->mutable_stream();
const size_t data_size = sizeof(int16_t) *
frame->samples_per_channel_ *
frame->num_channels_;
msg->set_input_data(frame->data_, data_size);
}
#endif
capture_audio_->DeinterleaveFrom(frame);
RETURN_ON_ERR(ProcessStreamLocked());
capture_audio_->InterleaveTo(frame, output_copy_needed(is_data_processed()));
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
if (debug_file_->Open()) {
audioproc::Stream* msg = event_msg_->mutable_stream();
const size_t data_size = sizeof(int16_t) *
frame->samples_per_channel_ *
frame->num_channels_;
msg->set_output_data(frame->data_, data_size);
RETURN_ON_ERR(WriteMessageToDebugFile());
}
#endif
return kNoError;
}
int AudioProcessingImpl::ProcessStreamLocked() {
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
if (debug_file_->Open()) {
audioproc::Stream* msg = event_msg_->mutable_stream();
msg->set_delay(stream_delay_ms_);
msg->set_drift(echo_cancellation_->stream_drift_samples());
msg->set_level(gain_control_->stream_analog_level());
msg->set_keypress(key_pressed_);
}
#endif
AudioBuffer* ca = capture_audio_.get(); // For brevity.
if (use_new_agc_ && gain_control_->is_enabled()) {
agc_manager_->AnalyzePreProcess(ca->data(0),
ca->num_channels(),
fwd_proc_format_.samples_per_channel());
}
bool data_processed = is_data_processed();
if (analysis_needed(data_processed)) {
ca->SplitIntoFrequencyBands();
}
#ifdef WEBRTC_BEAMFORMER
if (beamformer_enabled_) {
beamformer_->ProcessChunk(ca->split_channels_const_f(kBand0To8kHz),
ca->split_channels_const_f(kBand8To16kHz),
ca->num_channels(),
ca->samples_per_split_channel(),
ca->split_channels_f(kBand0To8kHz),
ca->split_channels_f(kBand8To16kHz));
ca->set_num_channels(1);
}
#endif
RETURN_ON_ERR(high_pass_filter_->ProcessCaptureAudio(ca));
RETURN_ON_ERR(gain_control_->AnalyzeCaptureAudio(ca));
RETURN_ON_ERR(noise_suppression_->AnalyzeCaptureAudio(ca));
RETURN_ON_ERR(echo_cancellation_->ProcessCaptureAudio(ca));
if (echo_control_mobile_->is_enabled() && noise_suppression_->is_enabled()) {
ca->CopyLowPassToReference();
}
RETURN_ON_ERR(noise_suppression_->ProcessCaptureAudio(ca));
RETURN_ON_ERR(echo_control_mobile_->ProcessCaptureAudio(ca));
RETURN_ON_ERR(voice_detection_->ProcessCaptureAudio(ca));
if (use_new_agc_ && gain_control_->is_enabled()) {
agc_manager_->Process(ca->split_bands_const(0)[kBand0To8kHz],
ca->samples_per_split_channel(),
split_rate_);
}
RETURN_ON_ERR(gain_control_->ProcessCaptureAudio(ca));
if (synthesis_needed(data_processed)) {
ca->MergeFrequencyBands();
}
// TODO(aluebs): Investigate if the transient suppression placement should be
// before or after the AGC.
if (transient_suppressor_enabled_) {
float voice_probability =
agc_manager_.get() ? agc_manager_->voice_probability() : 1.f;
transient_suppressor_->Suppress(ca->data_f(0),
ca->samples_per_channel(),
ca->num_channels(),
ca->split_bands_const_f(0)[kBand0To8kHz],
ca->samples_per_split_channel(),
ca->keyboard_data(),
ca->samples_per_keyboard_channel(),
voice_probability,
key_pressed_);
}
// The level estimator operates on the recombined data.
RETURN_ON_ERR(level_estimator_->ProcessStream(ca));
was_stream_delay_set_ = false;
return kNoError;
}
int AudioProcessingImpl::AnalyzeReverseStream(const float* const* data,
int samples_per_channel,
int sample_rate_hz,
ChannelLayout layout) {
CriticalSectionScoped crit_scoped(crit_);
if (data == NULL) {
return kNullPointerError;
}
const int num_channels = ChannelsFromLayout(layout);
RETURN_ON_ERR(MaybeInitializeLocked(fwd_in_format_.rate(),
fwd_out_format_.rate(),
sample_rate_hz,
fwd_in_format_.num_channels(),
fwd_out_format_.num_channels(),
num_channels));
if (samples_per_channel != rev_in_format_.samples_per_channel()) {
return kBadDataLengthError;
}
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
if (debug_file_->Open()) {
event_msg_->set_type(audioproc::Event::REVERSE_STREAM);
audioproc::ReverseStream* msg = event_msg_->mutable_reverse_stream();
const size_t channel_size =
sizeof(float) * rev_in_format_.samples_per_channel();
for (int i = 0; i < num_channels; ++i)
msg->add_channel(data[i], channel_size);
RETURN_ON_ERR(WriteMessageToDebugFile());
}
#endif
render_audio_->CopyFrom(data, samples_per_channel, layout);
return AnalyzeReverseStreamLocked();
}
int AudioProcessingImpl::AnalyzeReverseStream(AudioFrame* frame) {
CriticalSectionScoped crit_scoped(crit_);
if (frame == NULL) {
return kNullPointerError;
}
// Must be a native rate.
if (frame->sample_rate_hz_ != kSampleRate8kHz &&
frame->sample_rate_hz_ != kSampleRate16kHz &&
frame->sample_rate_hz_ != kSampleRate32kHz &&
frame->sample_rate_hz_ != kSampleRate48kHz) {
return kBadSampleRateError;
}
// This interface does not tolerate different forward and reverse rates.
if (frame->sample_rate_hz_ != fwd_in_format_.rate()) {
return kBadSampleRateError;
}
RETURN_ON_ERR(MaybeInitializeLocked(fwd_in_format_.rate(),
fwd_out_format_.rate(),
frame->sample_rate_hz_,
fwd_in_format_.num_channels(),
fwd_in_format_.num_channels(),
frame->num_channels_));
if (frame->samples_per_channel_ != rev_in_format_.samples_per_channel()) {
return kBadDataLengthError;
}
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
if (debug_file_->Open()) {
event_msg_->set_type(audioproc::Event::REVERSE_STREAM);
audioproc::ReverseStream* msg = event_msg_->mutable_reverse_stream();
const size_t data_size = sizeof(int16_t) *
frame->samples_per_channel_ *
frame->num_channels_;
msg->set_data(frame->data_, data_size);
RETURN_ON_ERR(WriteMessageToDebugFile());
}
#endif
render_audio_->DeinterleaveFrom(frame);
return AnalyzeReverseStreamLocked();
}
int AudioProcessingImpl::AnalyzeReverseStreamLocked() {
AudioBuffer* ra = render_audio_.get(); // For brevity.
if (rev_proc_format_.rate() == kSampleRate32kHz) {
ra->SplitIntoFrequencyBands();
}
RETURN_ON_ERR(echo_cancellation_->ProcessRenderAudio(ra));
RETURN_ON_ERR(echo_control_mobile_->ProcessRenderAudio(ra));
if (!use_new_agc_) {
RETURN_ON_ERR(gain_control_->ProcessRenderAudio(ra));
}
return kNoError;
}
int AudioProcessingImpl::set_stream_delay_ms(int delay) {
Error retval = kNoError;
was_stream_delay_set_ = true;
delay += delay_offset_ms_;
if (delay < 0) {
delay = 0;
retval = kBadStreamParameterWarning;
}
// TODO(ajm): the max is rather arbitrarily chosen; investigate.
if (delay > 500) {
delay = 500;
retval = kBadStreamParameterWarning;
}
stream_delay_ms_ = delay;
return retval;
}
int AudioProcessingImpl::stream_delay_ms() const {
return stream_delay_ms_;
}
bool AudioProcessingImpl::was_stream_delay_set() const {
return was_stream_delay_set_;
}
void AudioProcessingImpl::set_stream_key_pressed(bool key_pressed) {
key_pressed_ = key_pressed;
}
bool AudioProcessingImpl::stream_key_pressed() const {
return key_pressed_;
}
void AudioProcessingImpl::set_delay_offset_ms(int offset) {
CriticalSectionScoped crit_scoped(crit_);
delay_offset_ms_ = offset;
}
int AudioProcessingImpl::delay_offset_ms() const {
return delay_offset_ms_;
}
int AudioProcessingImpl::StartDebugRecording(
const char filename[AudioProcessing::kMaxFilenameSize]) {
CriticalSectionScoped crit_scoped(crit_);
assert(kMaxFilenameSize == FileWrapper::kMaxFileNameSize);
if (filename == NULL) {
return kNullPointerError;
}
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
// Stop any ongoing recording.
if (debug_file_->Open()) {
if (debug_file_->CloseFile() == -1) {
return kFileError;
}
}
if (debug_file_->OpenFile(filename, false) == -1) {
debug_file_->CloseFile();
return kFileError;
}
int err = WriteInitMessage();
if (err != kNoError) {
return err;
}
return kNoError;
#else
return kUnsupportedFunctionError;
#endif // WEBRTC_AUDIOPROC_DEBUG_DUMP
}
int AudioProcessingImpl::StartDebugRecording(FILE* handle) {
CriticalSectionScoped crit_scoped(crit_);
if (handle == NULL) {
return kNullPointerError;
}
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
// Stop any ongoing recording.
if (debug_file_->Open()) {
if (debug_file_->CloseFile() == -1) {
return kFileError;
}
}
if (debug_file_->OpenFromFileHandle(handle, true, false) == -1) {
return kFileError;
}
int err = WriteInitMessage();
if (err != kNoError) {
return err;
}
return kNoError;
#else
return kUnsupportedFunctionError;
#endif // WEBRTC_AUDIOPROC_DEBUG_DUMP
}
int AudioProcessingImpl::StartDebugRecordingForPlatformFile(
rtc::PlatformFile handle) {
FILE* stream = rtc::FdopenPlatformFileForWriting(handle);
return StartDebugRecording(stream);
}
int AudioProcessingImpl::StopDebugRecording() {
CriticalSectionScoped crit_scoped(crit_);
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
// We just return if recording hasn't started.
if (debug_file_->Open()) {
if (debug_file_->CloseFile() == -1) {
return kFileError;
}
}
return kNoError;
#else
return kUnsupportedFunctionError;
#endif // WEBRTC_AUDIOPROC_DEBUG_DUMP
}
EchoCancellation* AudioProcessingImpl::echo_cancellation() const {
return echo_cancellation_;
}
EchoControlMobile* AudioProcessingImpl::echo_control_mobile() const {
return echo_control_mobile_;
}
GainControl* AudioProcessingImpl::gain_control() const {
if (use_new_agc_) {
return gain_control_for_new_agc_.get();
}
return gain_control_;
}
HighPassFilter* AudioProcessingImpl::high_pass_filter() const {
return high_pass_filter_;
}
LevelEstimator* AudioProcessingImpl::level_estimator() const {
return level_estimator_;
}
NoiseSuppression* AudioProcessingImpl::noise_suppression() const {
return noise_suppression_;
}
VoiceDetection* AudioProcessingImpl::voice_detection() const {
return voice_detection_;
}
bool AudioProcessingImpl::is_data_processed() const {
if (beamformer_enabled_) {
return true;
}
int enabled_count = 0;
std::list<ProcessingComponent*>::const_iterator it;
for (it = component_list_.begin(); it != component_list_.end(); it++) {
if ((*it)->is_component_enabled()) {
enabled_count++;
}
}
// Data is unchanged if no components are enabled, or if only level_estimator_
// or voice_detection_ is enabled.
if (enabled_count == 0) {
return false;
} else if (enabled_count == 1) {
if (level_estimator_->is_enabled() || voice_detection_->is_enabled()) {
return false;
}
} else if (enabled_count == 2) {
if (level_estimator_->is_enabled() && voice_detection_->is_enabled()) {
return false;
}
}
return true;
}
bool AudioProcessingImpl::output_copy_needed(bool is_data_processed) const {
// Check if we've upmixed or downmixed the audio.
return ((fwd_out_format_.num_channels() != fwd_in_format_.num_channels()) ||
is_data_processed || transient_suppressor_enabled_);
}
bool AudioProcessingImpl::synthesis_needed(bool is_data_processed) const {
return (is_data_processed && (fwd_proc_format_.rate() == kSampleRate32kHz ||
fwd_proc_format_.rate() == kSampleRate48kHz));
}
bool AudioProcessingImpl::analysis_needed(bool is_data_processed) const {
if (!is_data_processed && !voice_detection_->is_enabled() &&
!transient_suppressor_enabled_) {
// Only level_estimator_ is enabled.
return false;
} else if (fwd_proc_format_.rate() == kSampleRate32kHz ||
fwd_proc_format_.rate() == kSampleRate48kHz) {
// Something besides level_estimator_ is enabled, and we have super-wb.
return true;
}
return false;
}
int AudioProcessingImpl::InitializeExperimentalAgc() {
if (use_new_agc_) {
if (!agc_manager_.get()) {
agc_manager_.reset(
new AgcManagerDirect(gain_control_, gain_control_for_new_agc_.get()));
}
agc_manager_->Initialize();
agc_manager_->SetCaptureMuted(output_will_be_muted_);
}
return kNoError;
}
int AudioProcessingImpl::InitializeTransient() {
if (transient_suppressor_enabled_) {
if (!transient_suppressor_.get()) {
transient_suppressor_.reset(new TransientSuppressor());
}
transient_suppressor_->Initialize(fwd_proc_format_.rate(),
split_rate_,
fwd_out_format_.num_channels());
}
return kNoError;
}
void AudioProcessingImpl::InitializeBeamformer() {
if (beamformer_enabled_) {
#ifdef WEBRTC_BEAMFORMER
beamformer_.reset(new Beamformer(kChunkSizeMs,
split_rate_,
array_geometry_));
#else
assert(false);
#endif
}
}
#ifdef WEBRTC_AUDIOPROC_DEBUG_DUMP
int AudioProcessingImpl::WriteMessageToDebugFile() {
int32_t size = event_msg_->ByteSize();
if (size <= 0) {
return kUnspecifiedError;
}
#if defined(WEBRTC_ARCH_BIG_ENDIAN)
// TODO(ajm): Use little-endian "on the wire". For the moment, we can be
// pretty safe in assuming little-endian.
#endif
if (!event_msg_->SerializeToString(&event_str_)) {
return kUnspecifiedError;
}
// Write message preceded by its size.
if (!debug_file_->Write(&size, sizeof(int32_t))) {
return kFileError;
}
if (!debug_file_->Write(event_str_.data(), event_str_.length())) {
return kFileError;
}
event_msg_->Clear();
return kNoError;
}
int AudioProcessingImpl::WriteInitMessage() {
event_msg_->set_type(audioproc::Event::INIT);
audioproc::Init* msg = event_msg_->mutable_init();
msg->set_sample_rate(fwd_in_format_.rate());
msg->set_num_input_channels(fwd_in_format_.num_channels());
msg->set_num_output_channels(fwd_out_format_.num_channels());
msg->set_num_reverse_channels(rev_in_format_.num_channels());
msg->set_reverse_sample_rate(rev_in_format_.rate());
msg->set_output_sample_rate(fwd_out_format_.rate());
int err = WriteMessageToDebugFile();
if (err != kNoError) {
return err;
}
return kNoError;
}
#endif // WEBRTC_AUDIOPROC_DEBUG_DUMP
} // namespace webrtc
|