New module for the adaptive level controlling functionality in the audio processing module

webrtc/modules/audio_processing/level_controller/level_controller.cc

Index: webrtc/modules/audio_processing/level_controller/level_controller.cc
diff --git a/webrtc/modules/audio_processing/level_controller/level_controller.cc b/webrtc/modules/audio_processing/level_controller/level_controller.cc
new file mode 100644
index 0000000000000000000000000000000000000000..fa26c059cbbb333c7fbe481b59ffb49495990bc3
--- /dev/null
+++ b/webrtc/modules/audio_processing/level_controller/level_controller.cc
@@ -0,0 +1,147 @@
+/*
+ *  Copyright (c) 2016 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/level_controller/level_controller.h"
+
+#include <algorithm>
+#include <numeric>
+
+#include "webrtc/base/array_view.h"
+#include "webrtc/base/checks.h"
+#include "webrtc/modules/audio_processing/audio_buffer.h"
+#include "webrtc/modules/audio_processing/level_controller/gain_applier.h"
+#include "webrtc/modules/audio_processing/level_controller/gain_selector.h"
+#include "webrtc/modules/audio_processing/level_controller/noise_level_estimator.h"
+#include "webrtc/modules/audio_processing/level_controller/peak_level_estimator.h"
+#include "webrtc/modules/audio_processing/level_controller/saturating_gain_estimator.h"
+#include "webrtc/modules/audio_processing/level_controller/signal_classifier.h"
+#include "webrtc/modules/audio_processing/logging/apm_data_dumper.h"
+
+namespace webrtc {
+namespace {
+
+void UpdateAndRemoveDcLevel(float forgetting_factor,
+                            float* dc_level,
+                            rtc::ArrayView<float> x) {
+  RTC_DCHECK(!x.empty());
+  float mean =
+      std::accumulate(x.begin(), x.end(), 0) / static_cast<float>(x.size());
+  *dc_level += forgetting_factor * (mean - *dc_level);
+
+  for (float& v : x) {
+    v -= *dc_level;
+  }
+}
+
+float FrameEnergy(const AudioBuffer& audio) {
+  float energy = 0.f;
+  for (size_t k = 0; k < audio.num_channels(); ++k) {
+    float channel_energy =
+        std::accumulate(audio.channels_const_f()[k],
+                        audio.channels_const_f()[k] + audio.num_frames(), 0,
+                        [](float a, float b) -> float { return a + b * b; });
+    energy = std::max(channel_energy, energy);
+  }
+  return energy;
+}
+
+float PeakLevel(const AudioBuffer& audio) {
+  float peak_level = 0.f;
+  for (size_t k = 0; k < audio.num_channels(); ++k) {
+    auto channel_peak_level = std::max_element(
+        audio.channels_const_f()[k],
+        audio.channels_const_f()[k] + audio.num_frames(),
+        [](float a, float b) { return std::abs(a) < std::abs(b); });
+    peak_level = std::max(*channel_peak_level, peak_level);
+  }
+  return peak_level;
+}
+
+}  // namespace
+
+int LevelController::instance_count_ = 0;
+
+LevelController::LevelController()
+    : data_dumper_(new ApmDataDumper(instance_count_)),
+      gain_applier_(data_dumper_.get()),
+      signal_classifier_(data_dumper_.get()) {
+  Initialize(AudioProcessing::kSampleRate48kHz);
+  ++instance_count_;
+}
+
+LevelController::~LevelController() {}
+
+void LevelController::Initialize(int sample_rate_hz) {
+  RTC_DCHECK(sample_rate_hz == AudioProcessing::kSampleRate8kHz ||
+             sample_rate_hz == AudioProcessing::kSampleRate16kHz ||
+             sample_rate_hz == AudioProcessing::kSampleRate32kHz ||
+             sample_rate_hz == AudioProcessing::kSampleRate48kHz);
+  data_dumper_->InitiateNewSetOfRecordings();
+  gain_selector_.Initialize(sample_rate_hz);
+  gain_applier_.Initialize(sample_rate_hz);
+  signal_classifier_.Initialize(sample_rate_hz);
+  noise_level_estimator_.Initialize(sample_rate_hz);
+  peak_level_estimator_.Initialize();
+  saturating_gain_estimator_.Initialize();
+
+  sample_rate_hz_ = rtc::Optional<int>(sample_rate_hz);
+  dc_forgetting_factor_ = 0.01f * sample_rate_hz / 48000.f;
+}
+
+void LevelController::Process(AudioBuffer* audio) {
+  RTC_DCHECK_LT(0u, audio->num_channels());
+  RTC_DCHECK_GE(2u, audio->num_channels());
+  RTC_DCHECK_NE(0.f, dc_forgetting_factor_);
+  RTC_DCHECK(sample_rate_hz_);
+  data_dumper_->DumpWav("lc_input", audio->num_frames(),
+                        audio->channels_const_f()[0], *sample_rate_hz_, 1);
+
+  // Remove DC level.
+  for (size_t k = 0; k < audio->num_channels(); ++k) {
+    UpdateAndRemoveDcLevel(
+        dc_forgetting_factor_, &dc_level_[k],
+        rtc::ArrayView<float>(audio->channels_f()[k], audio->num_frames()));
+  }
+
+  SignalClassifier::SignalType signal_type;
+  signal_classifier_.Analyze(*audio, &signal_type);
+  int tmp = static_cast<int>(signal_type);
+  data_dumper_->DumpRaw("lc_signal_type", 1, &tmp);
+
+  // Estimate the noise energy.
+  float noise_energy =
+      noise_level_estimator_.Analyze(signal_type, FrameEnergy(*audio));
+
+  // Estimate the overall signal peak level.
+  float peak_level =
+      peak_level_estimator_.Analyze(signal_type, PeakLevel(*audio));
+
+  float saturating_gain = saturating_gain_estimator_.GetGain();
+
+  // Compute the new gain to apply.
+  float new_gain = gain_selector_.GetNewGain(peak_level, noise_energy,
+                                             saturating_gain, signal_type);
+
+  // Apply the gain to the signal.
+  int num_saturations = gain_applier_.Process(new_gain, audio);
+
+  // Estimate the gain that saturates the overall signal.
+  saturating_gain_estimator_.Update(new_gain, num_saturations);
+
+  data_dumper_->DumpRaw("lc_selected_gain", 1, &new_gain);
+  data_dumper_->DumpRaw("lc_noise_energy", 1, &noise_energy);
+  data_dumper_->DumpRaw("lc_peak_level", 1, &peak_level);
+  data_dumper_->DumpRaw("lc_saturating_gain", 1, &saturating_gain);
+
+  data_dumper_->DumpWav("lc_output", audio->num_frames(),
+                        audio->channels_f()[0], *sample_rate_hz_, 1);
+}
+
+}  // namespace webrtc