media/base/audio_converter_unittest.cc

// Copyright (c) 2012 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.

// MSVC++ requires this to be set before any other includes to get M_PI.
#define _USE_MATH_DEFINES

#include <cmath>

#include "base/command_line.h"
#include "base/logging.h"
#include "base/memory/scoped_ptr.h"
#include "base/memory/scoped_vector.h"
#include "base/strings/string_number_conversions.h"
#include "base/time/time.h"
#include "media/base/audio_converter.h"
#include "media/base/fake_audio_render_callback.h"
#include "testing/gmock/include/gmock/gmock.h"
#include "testing/gtest/include/gtest/gtest.h"

namespace media {

// Command line switch for runtime adjustment of benchmark iterations.
static const char kBenchmarkIterations[] = "audio-converter-iterations";
static const int kDefaultIterations = 10;

// Parameters which control the many input case tests.
static const int kConvertInputs = 8;
static const int kConvertCycles = 3;

// Parameters used for testing.
static const int kBitsPerChannel = 32;
static const ChannelLayout kChannelLayout = CHANNEL_LAYOUT_STEREO;
static const int kHighLatencyBufferSize = 2048;
static const int kLowLatencyBufferSize = 256;
static const int kSampleRate = 48000;

// Number of full sine wave cycles for each Render() call.
static const int kSineCycles = 4;

// Tuple of <input rate, output rate, output channel layout, epsilon>.
typedef std::tr1::tuple<int, int, ChannelLayout, double> AudioConverterTestData;
class AudioConverterTest
    : public testing::TestWithParam<AudioConverterTestData> {
 public:
  AudioConverterTest()
      : epsilon_(std::tr1::get<3>(GetParam())) {
    // Create input and output parameters based on test parameters.
    input_parameters_ = AudioParameters(
        AudioParameters::AUDIO_PCM_LINEAR, kChannelLayout,
        std::tr1::get<0>(GetParam()), kBitsPerChannel, kHighLatencyBufferSize);
    output_parameters_ = AudioParameters(
        AudioParameters::AUDIO_PCM_LOW_LATENCY, std::tr1::get<2>(GetParam()),
        std::tr1::get<1>(GetParam()), 16, kLowLatencyBufferSize);

    converter_.reset(new AudioConverter(
        input_parameters_, output_parameters_, false));

    audio_bus_ = AudioBus::Create(output_parameters_);
    expected_audio_bus_ = AudioBus::Create(output_parameters_);

    // Allocate one callback for generating expected results.
    double step = kSineCycles / static_cast<double>(
        output_parameters_.frames_per_buffer());
    expected_callback_.reset(new FakeAudioRenderCallback(step));
  }

  // Creates |count| input callbacks to be used for conversion testing.
  void InitializeInputs(int count) {
    // Setup FakeAudioRenderCallback step to compensate for resampling.
    double scale_factor = input_parameters_.sample_rate() /
        static_cast<double>(output_parameters_.sample_rate());
    double step = kSineCycles / (scale_factor *
        static_cast<double>(output_parameters_.frames_per_buffer()));

    for (int i = 0; i < count; ++i) {
      fake_callbacks_.push_back(new FakeAudioRenderCallback(step));
      converter_->AddInput(fake_callbacks_[i]);
    }
  }

  // Resets all input callbacks to a pristine state.
  void Reset() {
    converter_->Reset();
    for (size_t i = 0; i < fake_callbacks_.size(); ++i)
      fake_callbacks_[i]->reset();
    expected_callback_->reset();
  }

  // Sets the volume on all input callbacks to |volume|.
  void SetVolume(float volume) {
    for (size_t i = 0; i < fake_callbacks_.size(); ++i)
      fake_callbacks_[i]->set_volume(volume);
  }

  // Validates audio data between |audio_bus_| and |expected_audio_bus_| from
  // |index|..|frames| after |scale| is applied to the expected audio data.
  bool ValidateAudioData(int index, int frames, float scale) {
    for (int i = 0; i < audio_bus_->channels(); ++i) {
      for (int j = index; j < frames; ++j) {
        double error = fabs(audio_bus_->channel(i)[j] -
            expected_audio_bus_->channel(i)[j] * scale);
        if (error > epsilon_) {
          EXPECT_NEAR(expected_audio_bus_->channel(i)[j] * scale,
                      audio_bus_->channel(i)[j], epsilon_)
              << " i=" << i << ", j=" << j;
          return false;
        }
      }
    }
    return true;
  }

  // Runs a single Convert() stage, fills |expected_audio_bus_| appropriately,
  // and validates equality with |audio_bus_| after |scale| is applied.
  bool RenderAndValidateAudioData(float scale) {
    // Render actual audio data.
    converter_->Convert(audio_bus_.get());

    // Render expected audio data.
    expected_callback_->Render(expected_audio_bus_.get(), 0);

    // Zero out unused channels in the expected AudioBus just as AudioConverter
    // would during channel mixing.
    for (int i = input_parameters_.channels();
         i < output_parameters_.channels(); ++i) {
      memset(expected_audio_bus_->channel(i), 0,
             audio_bus_->frames() * sizeof(*audio_bus_->channel(i)));
    }

    return ValidateAudioData(0, audio_bus_->frames(), scale);
  }

  // Fills |audio_bus_| fully with |value|.
  void FillAudioData(float value) {
    for (int i = 0; i < audio_bus_->channels(); ++i) {
      std::fill(audio_bus_->channel(i),
                audio_bus_->channel(i) + audio_bus_->frames(), value);
    }
  }

  // Verifies converter output with a |inputs| number of transform inputs.
  void RunTest(int inputs) {
    InitializeInputs(inputs);

    SetVolume(0);
    for (int i = 0; i < kConvertCycles; ++i)
      ASSERT_TRUE(RenderAndValidateAudioData(0));

    Reset();

    // Set a different volume for each input and verify the results.
    float total_scale = 0;
    for (size_t i = 0; i < fake_callbacks_.size(); ++i) {
      float volume = static_cast<float>(i) / fake_callbacks_.size();
      total_scale += volume;
      fake_callbacks_[i]->set_volume(volume);
    }
    for (int i = 0; i < kConvertCycles; ++i)
      ASSERT_TRUE(RenderAndValidateAudioData(total_scale));

    Reset();

    // Remove every other input.
    for (size_t i = 1; i < fake_callbacks_.size(); i += 2)
      converter_->RemoveInput(fake_callbacks_[i]);

    SetVolume(1);
    float scale = inputs > 1 ? inputs / 2.0f : inputs;
    for (int i = 0; i < kConvertCycles; ++i)
      ASSERT_TRUE(RenderAndValidateAudioData(scale));
  }

 protected:
  virtual ~AudioConverterTest() {}

  // Converter under test.
  scoped_ptr<AudioConverter> converter_;

  // Input and output parameters used for AudioConverter construction.
  AudioParameters input_parameters_;
  AudioParameters output_parameters_;

  // Destination AudioBus for AudioConverter output.
  scoped_ptr<AudioBus> audio_bus_;

  // AudioBus containing expected results for comparison with |audio_bus_|.
  scoped_ptr<AudioBus> expected_audio_bus_;

  // Vector of all input callbacks used to drive AudioConverter::Convert().
  ScopedVector<FakeAudioRenderCallback> fake_callbacks_;

  // Parallel input callback which generates the expected output.
  scoped_ptr<FakeAudioRenderCallback> expected_callback_;

  // Epsilon value with which to perform comparisons between |audio_bus_| and
  // |expected_audio_bus_|.
  double epsilon_;

  DISALLOW_COPY_AND_ASSIGN(AudioConverterTest);
};

// Ensure the buffer delay provided by AudioConverter is accurate.
TEST(AudioConverterTest, AudioDelay) {
  // Choose input and output parameters such that the transform must make
  // multiple calls to fill the buffer.
  AudioParameters input_parameters = AudioParameters(
      AudioParameters::AUDIO_PCM_LINEAR, kChannelLayout, kSampleRate,
      kBitsPerChannel, kLowLatencyBufferSize);
  AudioParameters output_parameters = AudioParameters(
      AudioParameters::AUDIO_PCM_LINEAR, kChannelLayout, kSampleRate * 2,
      kBitsPerChannel, kHighLatencyBufferSize);

  AudioConverter converter(input_parameters, output_parameters, false);
  FakeAudioRenderCallback callback(0.2);
  scoped_ptr<AudioBus> audio_bus = AudioBus::Create(output_parameters);
  converter.AddInput(&callback);
  converter.Convert(audio_bus.get());

  // Calculate the expected buffer delay for given AudioParameters.
  double input_sample_rate = input_parameters.sample_rate();
  int fill_count =
      (output_parameters.frames_per_buffer() * input_sample_rate /
       output_parameters.sample_rate()) / input_parameters.frames_per_buffer();

  base::TimeDelta input_frame_duration = base::TimeDelta::FromMicroseconds(
      base::Time::kMicrosecondsPerSecond / input_sample_rate);

  int expected_last_delay_milliseconds =
      fill_count * input_parameters.frames_per_buffer() *
      input_frame_duration.InMillisecondsF();

  EXPECT_EQ(expected_last_delay_milliseconds,
            callback.last_audio_delay_milliseconds());
}

// InputCallback that zero's out the provided AudioBus.  Used for benchmarking.
class NullInputProvider : public AudioConverter::InputCallback {
 public:
  NullInputProvider() {}
  virtual ~NullInputProvider() {}

  virtual double ProvideInput(AudioBus* audio_bus,
                              base::TimeDelta buffer_delay) OVERRIDE {
    audio_bus->Zero();
    return 1;
  }
};

// Benchmark for audio conversion.  Original benchmarks were run with
// --audio-converter-iterations=50000.
TEST(AudioConverterTest, ConvertBenchmark) {
  int benchmark_iterations = kDefaultIterations;
  std::string iterations(CommandLine::ForCurrentProcess()->GetSwitchValueASCII(
      kBenchmarkIterations));
  base::StringToInt(iterations, &benchmark_iterations);
  if (benchmark_iterations < kDefaultIterations)
    benchmark_iterations = kDefaultIterations;

  NullInputProvider fake_input1;
  NullInputProvider fake_input2;
  NullInputProvider fake_input3;

  printf("Benchmarking %d iterations:\n", benchmark_iterations);

  {
    // Create input and output parameters to convert between the two most common
    // sets of parameters (as indicated via UMA data).
    AudioParameters input_params(
        AudioParameters::AUDIO_PCM_LINEAR, CHANNEL_LAYOUT_MONO,
        48000, 16, 2048);
    AudioParameters output_params(
        AudioParameters::AUDIO_PCM_LINEAR, CHANNEL_LAYOUT_STEREO,
        44100, 16, 440);
    scoped_ptr<AudioBus> output_bus = AudioBus::Create(output_params);

    scoped_ptr<AudioConverter> converter(
        new AudioConverter(input_params, output_params, true));
    converter->AddInput(&fake_input1);
    converter->AddInput(&fake_input2);
    converter->AddInput(&fake_input3);

    // Benchmark Convert() w/ FIFO.
    base::TimeTicks start = base::TimeTicks::HighResNow();
    for (int i = 0; i < benchmark_iterations; ++i) {
      converter->Convert(output_bus.get());
    }
    double total_time_ms =
        (base::TimeTicks::HighResNow() - start).InMillisecondsF();
    printf("Convert() w/ Resampling took %.2fms.\n", total_time_ms);
  }

  // Create input and output parameters to convert between common buffer sizes
  // without any resampling for the FIFO vs no FIFO benchmarks.
  AudioParameters input_params(
      AudioParameters::AUDIO_PCM_LINEAR, CHANNEL_LAYOUT_STEREO,
      44100, 16, 2048);
  AudioParameters output_params(
      AudioParameters::AUDIO_PCM_LINEAR, CHANNEL_LAYOUT_STEREO,
      44100, 16, 440);
  scoped_ptr<AudioBus> output_bus = AudioBus::Create(output_params);

  {
    scoped_ptr<AudioConverter> converter(
        new AudioConverter(input_params, output_params, false));
    converter->AddInput(&fake_input1);
    converter->AddInput(&fake_input2);
    converter->AddInput(&fake_input3);

    // Benchmark Convert() w/ FIFO.
    base::TimeTicks start = base::TimeTicks::HighResNow();
    for (int i = 0; i < benchmark_iterations; ++i) {
      converter->Convert(output_bus.get());
    }
    double total_time_ms =
        (base::TimeTicks::HighResNow() - start).InMillisecondsF();
    printf("Convert() w/ FIFO took %.2fms.\n", total_time_ms);
  }

  {
    scoped_ptr<AudioConverter> converter(
        new AudioConverter(input_params, output_params, true));
    converter->AddInput(&fake_input1);
    converter->AddInput(&fake_input2);
    converter->AddInput(&fake_input3);

    // Benchmark Convert() w/o FIFO.
    base::TimeTicks start = base::TimeTicks::HighResNow();
    for (int i = 0; i < benchmark_iterations; ++i) {
      converter->Convert(output_bus.get());
    }
    double total_time_ms =
        (base::TimeTicks::HighResNow() - start).InMillisecondsF();
    printf("Convert() w/o FIFO took %.2fms.\n", total_time_ms);
  }
}

TEST_P(AudioConverterTest, NoInputs) {
  FillAudioData(1.0f);
  EXPECT_TRUE(RenderAndValidateAudioData(0.0f));
}

TEST_P(AudioConverterTest, OneInput) {
  RunTest(1);
}

TEST_P(AudioConverterTest, ManyInputs) {
  RunTest(kConvertInputs);
}

INSTANTIATE_TEST_CASE_P(
    AudioConverterTest, AudioConverterTest, testing::Values(
        // No resampling. No channel mixing.
        std::tr1::make_tuple(44100, 44100, CHANNEL_LAYOUT_STEREO, 0.00000048),

        // Upsampling. Channel upmixing.
        std::tr1::make_tuple(44100, 48000, CHANNEL_LAYOUT_QUAD, 0.033),

        // Downsampling. Channel downmixing.
        std::tr1::make_tuple(48000, 41000, CHANNEL_LAYOUT_MONO, 0.042)));

}  // namespace media