Added a bitexactness test for the intelligibility enhancer in the audio processing module

webrtc/modules/audio_processing/intelligibility_enhancer_unittest.cc

+/*
+ *  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 <vector>
+
+#include "testing/gtest/include/gtest/gtest.h"
+#include "webrtc/base/array_view.h"
+#include "webrtc/modules/audio_processing/audio_buffer.h"
+#include "webrtc/modules/audio_processing/intelligibility/intelligibility_enhancer.h"
+#include "webrtc/modules/audio_processing/noise_suppression_impl.h"
+#include "webrtc/modules/audio_processing/test/audio_buffer_tools.h"
+#include "webrtc/modules/audio_processing/test/bitexactness_tools.h"
+
+namespace webrtc {
+namespace {
+
+const size_t kNumFramesToProcess = 1000;
+
+int IntelligibilityEnhancerSampleRate(int sample_rate_hz) {
+  return (sample_rate_hz > AudioProcessing::kSampleRate16kHz
+              ? AudioProcessing::kSampleRate16kHz
+              : sample_rate_hz);
+}
+
+// Process one frame of data and produce the output.
+void ProcessOneFrame(int sample_rate_hz,
+                     AudioBuffer* render_audio_buffer,
+                     AudioBuffer* capture_audio_buffer,
+                     NoiseSuppressionImpl* noise_suppressor,
+                     IntelligibilityEnhancer* intelligibility_enhancer) {
+  if (sample_rate_hz > AudioProcessing::kSampleRate16kHz) {
+    render_audio_buffer->SplitIntoFrequencyBands();
+    capture_audio_buffer->SplitIntoFrequencyBands();
+  }
+
+  intelligibility_enhancer->ProcessRenderAudio(
+      render_audio_buffer->split_channels_f(kBand0To8kHz),
+      IntelligibilityEnhancerSampleRate(sample_rate_hz),
+      render_audio_buffer->num_channels());
+
+  noise_suppressor->AnalyzeCaptureAudio(capture_audio_buffer);
+  noise_suppressor->ProcessCaptureAudio(capture_audio_buffer);
+
+  intelligibility_enhancer->SetCaptureNoiseEstimate(
+      noise_suppressor->NoiseEstimate());
+
+  if (sample_rate_hz > AudioProcessing::kSampleRate16kHz) {
+    render_audio_buffer->MergeFrequencyBands();
+  }
+}
+
+// Processes a specified amount of frames, verifies the results and reports
+// any errors.
+void RunBitexactnessTest(int sample_rate_hz,
+                         size_t num_channels,
+                         rtc::ArrayView<const float> output_reference) {
+  const StreamConfig render_config(sample_rate_hz, num_channels, false);
+  AudioBuffer render_buffer(
+      render_config.num_frames(), render_config.num_channels(),
+      render_config.num_frames(), render_config.num_channels(),
+      render_config.num_frames());
+  test::InputAudioFile render_file(
+      test::GetApmRenderTestVectorFileName(sample_rate_hz));
+  std::vector<float> render_input(render_buffer.num_frames() *
+                                  render_buffer.num_channels());
+
+  const StreamConfig capture_config(sample_rate_hz, num_channels, false);
+  AudioBuffer capture_buffer(
+      capture_config.num_frames(), capture_config.num_channels(),
+      capture_config.num_frames(), capture_config.num_channels(),
+      capture_config.num_frames());
+  test::InputAudioFile capture_file(
+      test::GetApmCaptureTestVectorFileName(sample_rate_hz));
+  std::vector<float> capture_input(render_buffer.num_frames() *
+                                   capture_buffer.num_channels());
+
+  rtc::CriticalSection crit_capture;
+  NoiseSuppressionImpl noise_suppressor(&crit_capture);
+  noise_suppressor.Initialize(capture_config.num_channels(), sample_rate_hz);
+  noise_suppressor.Enable(true);
+
+  IntelligibilityEnhancer intelligibility_enhancer(
+      IntelligibilityEnhancerSampleRate(sample_rate_hz),
+      render_config.num_channels(), NoiseSuppressionImpl::num_noise_bins());
+
+  for (size_t frame_no = 0u; frame_no < kNumFramesToProcess; ++frame_no) {
+    ReadFloatSamplesFromStereoFile(render_buffer.num_frames(),
+                                   render_buffer.num_channels(), &render_file,
+                                   render_input);
+    ReadFloatSamplesFromStereoFile(render_buffer.num_frames(),

[aluebs-webrtc, 2016/03/22 11:53:29]

capture_buffer

[peah-webrtc, 2016/03/23 22:13:53]

Good catch!!!
Done.

+                                   capture_buffer.num_channels(), &capture_file,
+                                   capture_input);
+
+    test::CopyVectorToAudioBuffer(render_config, render_input, &render_buffer);
+    test::CopyVectorToAudioBuffer(capture_config, capture_input,
+                                  &capture_buffer);
+
+    ProcessOneFrame(sample_rate_hz, &render_buffer, &capture_buffer,
+                    &noise_suppressor, &intelligibility_enhancer);
+  }
+
+  // Extract and verify the test results.
+  std::vector<float> render_output;
+  test::ExtractVectorFromAudioBuffer(render_config, &render_buffer,
+                                     &render_output);
+
+  const float kTolerance = 1.f / static_cast<float>(1 << 15);
+
+  // Compare the output with the reference. Only the first values of the output
+  // from last frame processed are compared in order not having to specify all
+  // preceeding frames as testvectors. As the algorithm being tested has a
+  // memory, testing only the last frame implicitly also tests the preceeding
+  // frames.
+  EXPECT_TRUE(test::BitExactFrame(render_buffer.num_frames(),
+                                  render_config.num_channels(),
+                                  output_reference, render_output, kTolerance));
+}
+
+}  // namespace
+
+TEST(IntelligibilityEnhancerBitExactnessTest, Mono8kHz) {
+  const float kOutputReference[] = {-0.001892f, -0.003296f, -0.001953f};
+
+  RunBitexactnessTest(AudioProcessing::kSampleRate8kHz, 1, kOutputReference);
+}
+
+TEST(IntelligibilityEnhancerBitExactnessTest, Mono16kHz) {
+  const float kOutputReference[] = {-0.000977f, -0.003296f, -0.002441f};
+
+  RunBitexactnessTest(AudioProcessing::kSampleRate16kHz, 1, kOutputReference);
+}
+
+TEST(IntelligibilityEnhancerBitExactnessTest, Mono32kHz) {
+  const float kOutputReference[] = {0.003021f, -0.011780f, -0.008209f};
+
+  RunBitexactnessTest(AudioProcessing::kSampleRate32kHz, 1, kOutputReference);
+}
+
+TEST(IntelligibilityEnhancerBitExactnessTest, Mono48kHz) {
+  const float kOutputReference[] = {-0.027696f, -0.026253f, -0.018001f};
+
+  RunBitexactnessTest(AudioProcessing::kSampleRate48kHz, 1, kOutputReference);
+}
+
+TEST(IntelligibilityEnhancerBitExactnessTest, Stereo8kHz) {
+  const float kOutputReference[] = {0.021454f,  0.035919f, 0.026428f,
+                                    -0.000641f, 0.000366f, 0.000641f};
+
+  RunBitexactnessTest(AudioProcessing::kSampleRate8kHz, 2, kOutputReference);
+}
+
+TEST(IntelligibilityEnhancerBitExactnessTest, Stereo16kHz) {
+  const float kOutputReference[] = {0.021362f,  0.035736f,  0.023895f,
+                                    -0.001404f, -0.001465f, 0.000549f};
+
+  RunBitexactnessTest(AudioProcessing::kSampleRate16kHz, 2, kOutputReference);
+}
+
+TEST(IntelligibilityEnhancerBitExactnessTest, Stereo32kHz) {
+  const float kOutputReference[] = {0.030641f,  0.027406f,  0.028321f,
+                                    -0.001343f, -0.004578f, 0.000977f};
+
+  RunBitexactnessTest(AudioProcessing::kSampleRate32kHz, 2, kOutputReference);
+}
+
+TEST(IntelligibilityEnhancerBitExactnessTest, Stereo48kHz) {
+  const float kOutputReference[] = {-0.009276f, -0.001601f, -0.008255f,
+                                    -0.012975f, -0.015940f, -0.017820f};
+
+  RunBitexactnessTest(AudioProcessing::kSampleRate48kHz, 2, kOutputReference);
+}
+
+}  // namespace webrtc