WavReaderAdaptor is a simple adaptor of the existing class WavReader from webrtc/common_audio/wav_file.h. The adaptor was mainly needed to use dependency injection and easily test the MultiEndCall class (see https://codereview.webrtc.org/2761853002/).

The unit test ConversationalSpeechTest.MultiEndCallWavReaderAdaptorSine uses CreateSineWavFile() and writes temporary wav files that are used for test (deleted only if the test passes).

BUG=webrtc:7218

Review-Url: https://codereview.webrtc.org/2774423005
Cr-Commit-Position: refs/heads/master@{#17608}

webrtc/modules/audio_processing/test/conversational_speech/BUILD.gn

@@ -38,8 +38,6 @@ rtc_static_library("lib") {
     "timing.cc",
     "timing.h",
     "wavreader_abstract_factory.h",
-    "wavreader_adaptor.cc",
-    "wavreader_adaptor.h",
     "wavreader_factory.cc",
     "wavreader_factory.h",
     "wavreader_interface.h",
@@ -65,6 +63,7 @@ rtc_source_set("unittest") {
     ":lib",
     "../../../../../webrtc:webrtc_common",
     "../../../../../webrtc/base:rtc_base_approved",
+    "../../../../../webrtc/common_audio",
     "../../../../../webrtc/test:test_support",
     "//testing/gmock",
     "//testing/gtest",

webrtc/modules/audio_processing/test/conversational_speech/generator_unittest.cc

@@ -33,15 +33,22 @@
 // cases in which there are wrong offsets leading to self cross-talk (which is
 // rejected).
 
+// MSVC++ requires this to be set before any other includes to get M_PI.
+#define _USE_MATH_DEFINES
+
 #include <stdio.h>
+#include <cmath>
 #include <map>
 #include <memory>
 
 #include "webrtc/base/logging.h"
+#include "webrtc/base/pathutils.h"
+#include "webrtc/common_audio/wav_file.h"
 #include "webrtc/modules/audio_processing/test/conversational_speech/config.h"
 #include "webrtc/modules/audio_processing/test/conversational_speech/mock_wavreader_factory.h"
 #include "webrtc/modules/audio_processing/test/conversational_speech/multiend_call.h"
 #include "webrtc/modules/audio_processing/test/conversational_speech/timing.h"
+#include "webrtc/modules/audio_processing/test/conversational_speech/wavreader_factory.h"
 #include "webrtc/test/gmock.h"
 #include "webrtc/test/gtest.h"
 #include "webrtc/test/testsupport/fileutils.h"
@@ -55,7 +62,7 @@ using conversational_speech::SaveTiming;
 using conversational_speech::MockWavReaderFactory;
 using conversational_speech::MultiEndCall;
 using conversational_speech::Turn;
-using conversational_speech::WavReaderAbstractFactory;
+using conversational_speech::WavReaderFactory;
 
 const char* const audiotracks_path = "/path/to/audiotracks";
 const char* const timing_filepath = "/path/to/timing_file.txt";
@@ -89,6 +96,23 @@ std::unique_ptr<MockWavReaderFactory> CreateMockWavReaderFactory() {
                                kDefaultMockWavReaderFactoryParamsMap));
 }
 
+void CreateSineWavFile(const std::string& filepath,
+                       const MockWavReaderFactory::Params& params,
+                       float frequency = 440.0f) {
+  // Create samples.
+  constexpr double two_pi = 2.0 * M_PI;
+  std::vector<int16_t> samples(params.num_samples);
+  for (std::size_t i = 0; i < params.num_samples; ++i) {
+    // TODO(alessiob): the produced tone is not pure, improve.
+    samples[i] = std::lround(32767.0f * std::sin(
+        two_pi * i * frequency / params.sample_rate));
+  }
+
+  // Write samples.
+  WavWriter wav_writer(filepath, params.sample_rate, params.num_channels);
+  wav_writer.WriteSamples(samples.data(), params.num_samples);
+}
+
 }  // namespace
 
 using testing::_;
@@ -487,5 +511,34 @@ TEST_F(ConversationalSpeechTest, MultiEndCallSetupLongSequenceInvalid) {
   EXPECT_FALSE(multiend_call.valid());
 }
 
+TEST_F(ConversationalSpeechTest, MultiEndCallWavReaderAdaptorSine) {
+  // Parameters with which wav files are created.
+  constexpr int duration_seconds = 5;
+  const int sample_rates[] = {8000, 11025, 16000, 22050, 32000, 44100, 48000};
+
+  for (int sample_rate : sample_rates) {
+    const rtc::Pathname temp_filename(
+        OutputPath(), "TempSineWavFile_" + std::to_string(sample_rate)
+            + ".wav");
+
+    // Write wav file.
+    const std::size_t num_samples = duration_seconds * sample_rate;
+    MockWavReaderFactory::Params params = {sample_rate, 1u, num_samples};
+    CreateSineWavFile(temp_filename.pathname(), params);
+    LOG(LS_VERBOSE) << "wav file @" << sample_rate << " Hz created ("
+        << num_samples << " samples)";
+
+    // Load wav file and check if params match.
+    WavReaderFactory wav_reader_factory;
+    auto wav_reader = wav_reader_factory.Create(temp_filename.pathname());
+    EXPECT_EQ(sample_rate, wav_reader->SampleRate());
+    EXPECT_EQ(1u, wav_reader->NumChannels());
+    EXPECT_EQ(num_samples, wav_reader->NumSamples());
+
+    // Clean up.
+    remove(temp_filename.pathname().c_str());
+  }
+}
+
 }  // namespace test
 }  // namespace webrtc

webrtc/modules/audio_processing/test/conversational_speech/mock_wavreader.cc

@@ -20,9 +20,9 @@ MockWavReader::MockWavReader(
     int sample_rate, size_t num_channels, size_t num_samples)
       : sample_rate_(sample_rate), num_channels_(num_channels),
           num_samples_(num_samples) {
-  ON_CALL(*this, sample_rate()).WillByDefault(Return(sample_rate_));
-  ON_CALL(*this, num_channels()).WillByDefault(Return(num_channels_));
-  ON_CALL(*this, num_samples()).WillByDefault(Return(num_samples_));
+  ON_CALL(*this, SampleRate()).WillByDefault(Return(sample_rate_));
+  ON_CALL(*this, NumChannels()).WillByDefault(Return(num_channels_));
+  ON_CALL(*this, NumSamples()).WillByDefault(Return(num_samples_));
 }
 
 MockWavReader::~MockWavReader() = default;

webrtc/modules/audio_processing/test/conversational_speech/mock_wavreader.h

@@ -14,6 +14,7 @@
 #include <cstddef>
 #include <string>
 
+#include "webrtc/base/array_view.h"
 #include "webrtc/modules/audio_processing/test/conversational_speech/wavreader_interface.h"
 #include "webrtc/test/gmock.h"
 #include "webrtc/typedefs.h"
@@ -28,12 +29,12 @@ class MockWavReader : public WavReaderInterface {
   ~MockWavReader();
 
   // TODO(alessiob): use ON_CALL to return random samples.
-  MOCK_METHOD2(ReadFloatSamples, size_t(size_t, float*));
-  MOCK_METHOD2(ReadInt16Samples, size_t(size_t, int16_t*));
+  MOCK_METHOD1(ReadFloatSamples, size_t(rtc::ArrayView<float>));
+  MOCK_METHOD1(ReadInt16Samples, size_t(rtc::ArrayView<int16_t>));
 
-  MOCK_CONST_METHOD0(sample_rate, int());
-  MOCK_CONST_METHOD0(num_channels, size_t());
-  MOCK_CONST_METHOD0(num_samples, size_t());
+  MOCK_CONST_METHOD0(SampleRate, int());
+  MOCK_CONST_METHOD0(NumChannels, size_t());
+  MOCK_CONST_METHOD0(NumSamples, size_t());
 
  private:
   const int sample_rate_;

webrtc/modules/audio_processing/test/conversational_speech/multiend_call.cc

@@ -120,9 +120,9 @@ bool MultiEndCall::CheckTiming() {
 
     // Begin and end timestamps for the current turn.
     int offset_samples = millisecond_to_samples(
-        turn.offset, it->second->sample_rate());
-    size_t begin_timestamp = last_turn.end + offset_samples;
-    size_t end_timestamp = begin_timestamp + it->second->num_samples();
+        turn.offset, it->second->SampleRate());
+    std::size_t begin_timestamp = last_turn.end + offset_samples;
+    std::size_t end_timestamp = begin_timestamp + it->second->NumSamples();
     LOG(LS_INFO) << "turn #" << turn_index << " " << begin_timestamp
         << "-" << end_timestamp << " ms";
 

webrtc/modules/audio_processing/test/conversational_speech/wavreader_adaptor.cc

@@ -1,58 +0,0 @@
-/*
- *  Copyright (c) 2017 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/test/conversational_speech/wavreader_adaptor.h"
-
-#include "webrtc/base/checks.h"
-
-namespace webrtc {
-namespace test {
-namespace conversational_speech {
-
-WavReaderAdaptor::WavReaderAdaptor(const std::string& filepath) {
-  // TODO(alessiob): implement.
-}
-
-WavReaderAdaptor::~WavReaderAdaptor() {}
-
-size_t WavReaderAdaptor::ReadFloatSamples(size_t num_samples, float* samples) {
-  // TODO(alessiob): implement.
-  FATAL();
-  return 0u;
-}
-
-size_t WavReaderAdaptor::ReadInt16Samples(
-    size_t num_samples, int16_t* samples) {
-  // TODO(alessiob): implement.
-  FATAL();
-  return 0u;
-}
-
-int WavReaderAdaptor::sample_rate() const {
-  // TODO(alessiob): implement.
-  FATAL();
-  return 0;
-}
-
-size_t WavReaderAdaptor::num_channels() const {
-  // TODO(alessiob): implement.
-  FATAL();
-  return 0u;
-}
-
-size_t WavReaderAdaptor::num_samples() const {
-  // TODO(alessiob): implement.
-  FATAL();
-  return 0u;
-}
-
-}  // namespace conversational_speech
-}  // namespace test
-}  // namespace webrtc

webrtc/modules/audio_processing/test/conversational_speech/wavreader_adaptor.h

@@ -1,42 +0,0 @@
-/*
- *  Copyright (c) 2017 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.
- */
-
-#ifndef WEBRTC_MODULES_AUDIO_PROCESSING_TEST_CONVERSATIONAL_SPEECH_WAVREADER_ADAPTOR_H_
-#define WEBRTC_MODULES_AUDIO_PROCESSING_TEST_CONVERSATIONAL_SPEECH_WAVREADER_ADAPTOR_H_
-
-#include "webrtc/modules/audio_processing/test/conversational_speech/wavreader_interface.h"
-
-#include <cstddef>
-#include <string>
-
-#include "webrtc/typedefs.h"
-
-namespace webrtc {
-namespace test {
-namespace conversational_speech {
-
-class WavReaderAdaptor : public WavReaderInterface {
- public:
-  explicit WavReaderAdaptor(const std::string& filepath);
-  ~WavReaderAdaptor() override;
-
-  size_t ReadFloatSamples(size_t num_samples, float* samples) override;
-  size_t ReadInt16Samples(size_t num_samples, int16_t* samples) override;
-
-  int sample_rate() const override;
-  size_t num_channels() const override;
-  size_t num_samples() const override;
-};
-
-}  // namespace conversational_speech
-}  // namespace test
-}  // namespace webrtc
-
-#endif  // WEBRTC_MODULES_AUDIO_PROCESSING_TEST_CONVERSATIONAL_SPEECH_WAVREADER_ADAPTOR_H_

webrtc/modules/audio_processing/test/conversational_speech/wavreader_factory.cc

@@ -10,10 +10,51 @@
 
 #include "webrtc/modules/audio_processing/test/conversational_speech/wavreader_factory.h"
 
-#include "webrtc/modules/audio_processing/test/conversational_speech/wavreader_adaptor.h"
+#include <cstddef>
+
+#include "webrtc/base/array_view.h"
+#include "webrtc/base/checks.h"
+#include "webrtc/common_audio/wav_file.h"
+#include "webrtc/typedefs.h"
 
 namespace webrtc {
 namespace test {
+namespace {
+
+using conversational_speech::WavReaderInterface;
+
+class WavReaderAdaptor final : public WavReaderInterface {
+ public:
+  explicit WavReaderAdaptor(const std::string& filepath)
+      : wav_reader_(filepath) {}
+  ~WavReaderAdaptor() override = default;
+
+  size_t ReadFloatSamples(rtc::ArrayView<float> samples) override {
+    return wav_reader_.ReadSamples(samples.size(), samples.begin());
+  }
+
+  size_t ReadInt16Samples(rtc::ArrayView<int16_t> samples) override {
+    return wav_reader_.ReadSamples(samples.size(), samples.begin());
+  }
+
+  int SampleRate() const override {
+    return wav_reader_.sample_rate();
+  }
+
+  size_t NumChannels() const override {
+    return wav_reader_.num_channels();
+  }
+
+  size_t NumSamples() const override {
+    return wav_reader_.num_samples();
+  }
+
+ private:
+  WavReader wav_reader_;
+};
+
+}  // namespace
+
 namespace conversational_speech {
 
 WavReaderFactory::WavReaderFactory() = default;

webrtc/modules/audio_processing/test/conversational_speech/wavreader_interface.h

@@ -13,6 +13,7 @@
 
 #include <stddef.h>
 
+#include "webrtc/base/array_view.h"
 #include "webrtc/typedefs.h"
 
 namespace webrtc {
@@ -24,13 +25,13 @@ class WavReaderInterface {
   virtual ~WavReaderInterface() = default;
 
   // Returns the number of samples read.
-  virtual size_t ReadFloatSamples(size_t num_samples, float* samples) = 0;
-  virtual size_t ReadInt16Samples(size_t num_samples, int16_t* samples) = 0;
+  virtual size_t ReadFloatSamples(rtc::ArrayView<float> samples) = 0;
+  virtual size_t ReadInt16Samples(rtc::ArrayView<int16_t> samples) = 0;
 
   // Getters.
-  virtual int sample_rate() const = 0;
-  virtual size_t num_channels() const = 0;
-  virtual size_t num_samples() const = 0;
+  virtual int SampleRate() const = 0;
+  virtual size_t NumChannels() const = 0;
+  virtual size_t NumSamples() const = 0;
 };
 
 }  // namespace conversational_speech