blob: 20b7d299f26d1806c70e3aa557e82b13a5caf118 [file] [log] [blame]
/*
* 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 "modules/audio_mixer/audio_mixer_impl.h"
#include <string.h>
#include <cstdint>
#include <limits>
#include <memory>
#include <string>
#include <utility>
#include <vector>
#include "absl/types/optional.h"
#include "api/audio/audio_mixer.h"
#include "api/rtp_packet_info.h"
#include "api/rtp_packet_infos.h"
#include "api/units/timestamp.h"
#include "modules/audio_mixer/default_output_rate_calculator.h"
#include "rtc_base/checks.h"
#include "rtc_base/strings/string_builder.h"
#include "rtc_base/task_queue_for_test.h"
#include "test/gmock.h"
#include "test/gtest.h"
using ::testing::_;
using ::testing::Exactly;
using ::testing::Invoke;
using ::testing::Return;
using ::testing::UnorderedElementsAre;
namespace webrtc {
namespace {
constexpr int kDefaultSampleRateHz = 48000;
// Utility function that resets the frame member variables with
// sensible defaults.
void ResetFrame(AudioFrame* frame) {
frame->sample_rate_hz_ = kDefaultSampleRateHz;
frame->num_channels_ = 1;
// Frame duration 10ms.
frame->samples_per_channel_ = kDefaultSampleRateHz / 100;
frame->vad_activity_ = AudioFrame::kVadActive;
frame->speech_type_ = AudioFrame::kNormalSpeech;
}
std::string ProduceDebugText(int sample_rate_hz,
int number_of_channels,
int number_of_sources) {
rtc::StringBuilder ss;
ss << "Sample rate: " << sample_rate_hz << " ";
ss << "Number of channels: " << number_of_channels << " ";
ss << "Number of sources: " << number_of_sources;
return ss.Release();
}
AudioFrame frame_for_mixing;
} // namespace
class MockMixerAudioSource : public ::testing::NiceMock<AudioMixer::Source> {
public:
MockMixerAudioSource()
: fake_audio_frame_info_(AudioMixer::Source::AudioFrameInfo::kNormal) {
ON_CALL(*this, GetAudioFrameWithInfo(_, _))
.WillByDefault(
Invoke(this, &MockMixerAudioSource::FakeAudioFrameWithInfo));
ON_CALL(*this, PreferredSampleRate())
.WillByDefault(Return(kDefaultSampleRateHz));
}
MOCK_METHOD(AudioFrameInfo,
GetAudioFrameWithInfo,
(int sample_rate_hz, AudioFrame* audio_frame),
(override));
MOCK_METHOD(int, PreferredSampleRate, (), (const, override));
MOCK_METHOD(int, Ssrc, (), (const, override));
AudioFrame* fake_frame() { return &fake_frame_; }
AudioFrameInfo fake_info() { return fake_audio_frame_info_; }
void set_fake_info(const AudioFrameInfo audio_frame_info) {
fake_audio_frame_info_ = audio_frame_info;
}
void set_packet_infos(const RtpPacketInfos& packet_infos) {
packet_infos_ = packet_infos;
}
private:
AudioFrameInfo FakeAudioFrameWithInfo(int sample_rate_hz,
AudioFrame* audio_frame) {
audio_frame->CopyFrom(fake_frame_);
audio_frame->sample_rate_hz_ = sample_rate_hz;
audio_frame->samples_per_channel_ =
rtc::CheckedDivExact(sample_rate_hz, 100);
audio_frame->packet_infos_ = packet_infos_;
return fake_info();
}
AudioFrame fake_frame_;
AudioFrameInfo fake_audio_frame_info_;
RtpPacketInfos packet_infos_;
};
class CustomRateCalculator : public OutputRateCalculator {
public:
explicit CustomRateCalculator(int rate) : rate_(rate) {}
int CalculateOutputRateFromRange(
rtc::ArrayView<const int> preferred_rates) override {
return rate_;
}
private:
const int rate_;
};
// Creates participants from `frames` and `frame_info` and adds them
// to the mixer. Compares mixed status with `expected_status`
void MixAndCompare(
const std::vector<AudioFrame>& frames,
const std::vector<AudioMixer::Source::AudioFrameInfo>& frame_info,
const std::vector<bool>& expected_status) {
const size_t num_audio_sources = frames.size();
RTC_DCHECK(frames.size() == frame_info.size());
RTC_DCHECK(frame_info.size() == expected_status.size());
const auto mixer = AudioMixerImpl::Create();
std::vector<MockMixerAudioSource> participants(num_audio_sources);
for (size_t i = 0; i < num_audio_sources; ++i) {
participants[i].fake_frame()->CopyFrom(frames[i]);
participants[i].set_fake_info(frame_info[i]);
}
for (size_t i = 0; i < num_audio_sources; ++i) {
EXPECT_TRUE(mixer->AddSource(&participants[i]));
EXPECT_CALL(participants[i], GetAudioFrameWithInfo(kDefaultSampleRateHz, _))
.Times(Exactly(1));
}
mixer->Mix(1, &frame_for_mixing);
for (size_t i = 0; i < num_audio_sources; ++i) {
EXPECT_EQ(expected_status[i],
mixer->GetAudioSourceMixabilityStatusForTest(&participants[i]))
<< "Mixed status of AudioSource #" << i << " wrong.";
}
}
void MixMonoAtGivenNativeRate(int native_sample_rate,
AudioFrame* mix_frame,
rtc::scoped_refptr<AudioMixer> mixer,
MockMixerAudioSource* audio_source) {
ON_CALL(*audio_source, PreferredSampleRate())
.WillByDefault(Return(native_sample_rate));
audio_source->fake_frame()->sample_rate_hz_ = native_sample_rate;
audio_source->fake_frame()->samples_per_channel_ = native_sample_rate / 100;
mixer->Mix(1, mix_frame);
}
TEST(AudioMixer, LargestEnergyVadActiveMixed) {
constexpr int kAudioSources =
AudioMixerImpl::kDefaultNumberOfMixedAudioSources + 3;
const auto mixer = AudioMixerImpl::Create();
MockMixerAudioSource participants[kAudioSources];
for (int i = 0; i < kAudioSources; ++i) {
ResetFrame(participants[i].fake_frame());
// We set the 80-th sample value since the first 80 samples may be
// modified by a ramped-in window.
participants[i].fake_frame()->mutable_data()[80] = i;
EXPECT_TRUE(mixer->AddSource(&participants[i]));
EXPECT_CALL(participants[i], GetAudioFrameWithInfo(_, _)).Times(Exactly(1));
}
// Last participant gives audio frame with passive VAD, although it has the
// largest energy.
participants[kAudioSources - 1].fake_frame()->vad_activity_ =
AudioFrame::kVadPassive;
AudioFrame audio_frame;
mixer->Mix(1, // number of channels
&audio_frame);
for (int i = 0; i < kAudioSources; ++i) {
bool is_mixed =
mixer->GetAudioSourceMixabilityStatusForTest(&participants[i]);
if (i == kAudioSources - 1 ||
i < kAudioSources - 1 -
AudioMixerImpl::kDefaultNumberOfMixedAudioSources) {
EXPECT_FALSE(is_mixed)
<< "Mixing status of AudioSource #" << i << " wrong.";
} else {
EXPECT_TRUE(is_mixed)
<< "Mixing status of AudioSource #" << i << " wrong.";
}
}
}
TEST(AudioMixer, FrameNotModifiedForSingleParticipant) {
const auto mixer = AudioMixerImpl::Create();
MockMixerAudioSource participant;
ResetFrame(participant.fake_frame());
const size_t n_samples = participant.fake_frame()->samples_per_channel_;
// Modify the frame so that it's not zero.
int16_t* fake_frame_data = participant.fake_frame()->mutable_data();
for (size_t j = 0; j < n_samples; ++j) {
fake_frame_data[j] = static_cast<int16_t>(j);
}
EXPECT_TRUE(mixer->AddSource(&participant));
EXPECT_CALL(participant, GetAudioFrameWithInfo(_, _)).Times(Exactly(2));
AudioFrame audio_frame;
// Two mix iteration to compare after the ramp-up step.
for (int i = 0; i < 2; ++i) {
mixer->Mix(1, // number of channels
&audio_frame);
}
EXPECT_EQ(0, memcmp(participant.fake_frame()->data(), audio_frame.data(),
n_samples));
}
TEST(AudioMixer, SourceAtNativeRateShouldNeverResample) {
const auto mixer = AudioMixerImpl::Create();
MockMixerAudioSource audio_source;
ResetFrame(audio_source.fake_frame());
mixer->AddSource(&audio_source);
for (auto frequency : {8000, 16000, 32000, 48000}) {
EXPECT_CALL(audio_source, GetAudioFrameWithInfo(frequency, _))
.Times(Exactly(1));
MixMonoAtGivenNativeRate(frequency, &frame_for_mixing, mixer,
&audio_source);
}
}
TEST(AudioMixer, MixerShouldMixAtNativeSourceRate) {
const auto mixer = AudioMixerImpl::Create();
MockMixerAudioSource audio_source;
ResetFrame(audio_source.fake_frame());
mixer->AddSource(&audio_source);
for (auto frequency : {8000, 16000, 32000, 48000}) {
MixMonoAtGivenNativeRate(frequency, &frame_for_mixing, mixer,
&audio_source);
EXPECT_EQ(frequency, frame_for_mixing.sample_rate_hz_);
}
}
TEST(AudioMixer, MixerShouldAlwaysMixAtNativeRate) {
const auto mixer = AudioMixerImpl::Create();
MockMixerAudioSource participant;
ResetFrame(participant.fake_frame());
mixer->AddSource(&participant);
const int needed_frequency = 44100;
ON_CALL(participant, PreferredSampleRate())
.WillByDefault(Return(needed_frequency));
// We expect mixing frequency to be native and >= needed_frequency.
const int expected_mix_frequency = 48000;
EXPECT_CALL(participant, GetAudioFrameWithInfo(expected_mix_frequency, _))
.Times(Exactly(1));
participant.fake_frame()->sample_rate_hz_ = expected_mix_frequency;
participant.fake_frame()->samples_per_channel_ = expected_mix_frequency / 100;
mixer->Mix(1, &frame_for_mixing);
EXPECT_EQ(48000, frame_for_mixing.sample_rate_hz_);
}
// Check that the mixing rate is always >= participants preferred rate.
TEST(AudioMixer, ShouldNotCauseQualityLossForMultipleSources) {
const auto mixer = AudioMixerImpl::Create();
std::vector<MockMixerAudioSource> audio_sources(2);
const std::vector<int> source_sample_rates = {8000, 16000};
for (int i = 0; i < 2; ++i) {
auto& source = audio_sources[i];
ResetFrame(source.fake_frame());
mixer->AddSource(&source);
const auto sample_rate = source_sample_rates[i];
EXPECT_CALL(source, PreferredSampleRate()).WillOnce(Return(sample_rate));
EXPECT_CALL(source, GetAudioFrameWithInfo(::testing::Ge(sample_rate), _));
}
mixer->Mix(1, &frame_for_mixing);
}
TEST(AudioMixer, ParticipantNumberOfChannels) {
const auto mixer = AudioMixerImpl::Create();
MockMixerAudioSource participant;
ResetFrame(participant.fake_frame());
EXPECT_TRUE(mixer->AddSource(&participant));
for (size_t number_of_channels : {1, 2}) {
EXPECT_CALL(participant, GetAudioFrameWithInfo(kDefaultSampleRateHz, _))
.Times(Exactly(1));
mixer->Mix(number_of_channels, &frame_for_mixing);
EXPECT_EQ(number_of_channels, frame_for_mixing.num_channels_);
}
}
// Maximal amount of participants are mixed one iteration, then
// another participant with higher energy is added.
TEST(AudioMixer, RampedOutSourcesShouldNotBeMarkedMixed) {
constexpr int kAudioSources =
AudioMixerImpl::kDefaultNumberOfMixedAudioSources + 1;
const auto mixer = AudioMixerImpl::Create();
MockMixerAudioSource participants[kAudioSources];
for (int i = 0; i < kAudioSources; ++i) {
ResetFrame(participants[i].fake_frame());
// Set the participant audio energy to increase with the index
// `i`.
participants[i].fake_frame()->mutable_data()[0] = 100 * i;
}
// Add all participants but the loudest for mixing.
for (int i = 0; i < kAudioSources - 1; ++i) {
EXPECT_TRUE(mixer->AddSource(&participants[i]));
EXPECT_CALL(participants[i], GetAudioFrameWithInfo(kDefaultSampleRateHz, _))
.Times(Exactly(1));
}
// First mixer iteration
mixer->Mix(1, &frame_for_mixing);
// All participants but the loudest should have been mixed.
for (int i = 0; i < kAudioSources - 1; ++i) {
EXPECT_TRUE(mixer->GetAudioSourceMixabilityStatusForTest(&participants[i]))
<< "Mixed status of AudioSource #" << i << " wrong.";
}
// Add new participant with higher energy.
EXPECT_TRUE(mixer->AddSource(&participants[kAudioSources - 1]));
for (int i = 0; i < kAudioSources; ++i) {
EXPECT_CALL(participants[i], GetAudioFrameWithInfo(kDefaultSampleRateHz, _))
.Times(Exactly(1));
}
mixer->Mix(1, &frame_for_mixing);
// The most quiet participant should not have been mixed.
EXPECT_FALSE(mixer->GetAudioSourceMixabilityStatusForTest(&participants[0]))
<< "Mixed status of AudioSource #0 wrong.";
// The loudest participants should have been mixed.
for (int i = 1; i < kAudioSources; ++i) {
EXPECT_EQ(true,
mixer->GetAudioSourceMixabilityStatusForTest(&participants[i]))
<< "Mixed status of AudioSource #" << i << " wrong.";
}
}
// This test checks that the initialization and participant addition
// can be done on a different thread.
TEST(AudioMixer, ConstructFromOtherThread) {
TaskQueueForTest init_queue("init");
rtc::scoped_refptr<AudioMixer> mixer;
init_queue.SendTask([&mixer]() { mixer = AudioMixerImpl::Create(); });
MockMixerAudioSource participant;
EXPECT_CALL(participant, PreferredSampleRate())
.WillRepeatedly(Return(kDefaultSampleRateHz));
ResetFrame(participant.fake_frame());
TaskQueueForTest participant_queue("participant");
participant_queue.SendTask(
[&mixer, &participant]() { mixer->AddSource(&participant); });
EXPECT_CALL(participant, GetAudioFrameWithInfo(kDefaultSampleRateHz, _))
.Times(Exactly(1));
// Do one mixer iteration
mixer->Mix(1, &frame_for_mixing);
}
TEST(AudioMixer, MutedShouldMixAfterUnmuted) {
constexpr int kAudioSources =
AudioMixerImpl::kDefaultNumberOfMixedAudioSources + 1;
std::vector<AudioFrame> frames(kAudioSources);
for (auto& frame : frames) {
ResetFrame(&frame);
}
std::vector<AudioMixer::Source::AudioFrameInfo> frame_info(
kAudioSources, AudioMixer::Source::AudioFrameInfo::kNormal);
frame_info[0] = AudioMixer::Source::AudioFrameInfo::kMuted;
std::vector<bool> expected_status(kAudioSources, true);
expected_status[0] = false;
MixAndCompare(frames, frame_info, expected_status);
}
TEST(AudioMixer, PassiveShouldMixAfterNormal) {
constexpr int kAudioSources =
AudioMixerImpl::kDefaultNumberOfMixedAudioSources + 1;
std::vector<AudioFrame> frames(kAudioSources);
for (auto& frame : frames) {
ResetFrame(&frame);
}
std::vector<AudioMixer::Source::AudioFrameInfo> frame_info(
kAudioSources, AudioMixer::Source::AudioFrameInfo::kNormal);
frames[0].vad_activity_ = AudioFrame::kVadPassive;
std::vector<bool> expected_status(kAudioSources, true);
expected_status[0] = false;
MixAndCompare(frames, frame_info, expected_status);
}
TEST(AudioMixer, ActiveShouldMixBeforeLoud) {
constexpr int kAudioSources =
AudioMixerImpl::kDefaultNumberOfMixedAudioSources + 1;
std::vector<AudioFrame> frames(kAudioSources);
for (auto& frame : frames) {
ResetFrame(&frame);
}
std::vector<AudioMixer::Source::AudioFrameInfo> frame_info(
kAudioSources, AudioMixer::Source::AudioFrameInfo::kNormal);
frames[0].vad_activity_ = AudioFrame::kVadPassive;
int16_t* frame_data = frames[0].mutable_data();
std::fill(frame_data, frame_data + kDefaultSampleRateHz / 100,
std::numeric_limits<int16_t>::max());
std::vector<bool> expected_status(kAudioSources, true);
expected_status[0] = false;
MixAndCompare(frames, frame_info, expected_status);
}
TEST(AudioMixer, ShouldMixUpToSpecifiedNumberOfSourcesToMix) {
constexpr int kAudioSources = 5;
constexpr int kSourcesToMix = 2;
std::vector<AudioFrame> frames(kAudioSources);
for (auto& frame : frames) {
ResetFrame(&frame);
}
std::vector<AudioMixer::Source::AudioFrameInfo> frame_info(
kAudioSources, AudioMixer::Source::AudioFrameInfo::kNormal);
// Set up to kSourceToMix sources with kVadActive so that they're mixed.
const std::vector<AudioFrame::VADActivity> kVadActivities = {
AudioFrame::kVadUnknown, AudioFrame::kVadPassive, AudioFrame::kVadPassive,
AudioFrame::kVadActive, AudioFrame::kVadActive};
// Populate VAD and frame for all sources.
for (int i = 0; i < kAudioSources; i++) {
frames[i].vad_activity_ = kVadActivities[i];
}
std::vector<MockMixerAudioSource> participants(kAudioSources);
for (int i = 0; i < kAudioSources; ++i) {
participants[i].fake_frame()->CopyFrom(frames[i]);
participants[i].set_fake_info(frame_info[i]);
}
const auto mixer = AudioMixerImpl::Create(kSourcesToMix);
for (int i = 0; i < kAudioSources; ++i) {
EXPECT_TRUE(mixer->AddSource(&participants[i]));
EXPECT_CALL(participants[i], GetAudioFrameWithInfo(kDefaultSampleRateHz, _))
.Times(Exactly(1));
}
mixer->Mix(1, &frame_for_mixing);
std::vector<bool> expected_status = {false, false, false, true, true};
for (int i = 0; i < kAudioSources; ++i) {
EXPECT_EQ(expected_status[i],
mixer->GetAudioSourceMixabilityStatusForTest(&participants[i]))
<< "Wrong mix status for source #" << i << " is wrong";
}
}
TEST(AudioMixer, UnmutedShouldMixBeforeLoud) {
constexpr int kAudioSources =
AudioMixerImpl::kDefaultNumberOfMixedAudioSources + 1;
std::vector<AudioFrame> frames(kAudioSources);
for (auto& frame : frames) {
ResetFrame(&frame);
}
std::vector<AudioMixer::Source::AudioFrameInfo> frame_info(
kAudioSources, AudioMixer::Source::AudioFrameInfo::kNormal);
frame_info[0] = AudioMixer::Source::AudioFrameInfo::kMuted;
int16_t* frame_data = frames[0].mutable_data();
std::fill(frame_data, frame_data + kDefaultSampleRateHz / 100,
std::numeric_limits<int16_t>::max());
std::vector<bool> expected_status(kAudioSources, true);
expected_status[0] = false;
MixAndCompare(frames, frame_info, expected_status);
}
TEST(AudioMixer, MixingRateShouldBeDecidedByRateCalculator) {
constexpr int kOutputRate = 22000;
const auto mixer =
AudioMixerImpl::Create(std::unique_ptr<OutputRateCalculator>(
new CustomRateCalculator(kOutputRate)),
true);
MockMixerAudioSource audio_source;
mixer->AddSource(&audio_source);
ResetFrame(audio_source.fake_frame());
EXPECT_CALL(audio_source, GetAudioFrameWithInfo(kOutputRate, _))
.Times(Exactly(1));
mixer->Mix(1, &frame_for_mixing);
}
TEST(AudioMixer, ZeroSourceRateShouldBeDecidedByRateCalculator) {
constexpr int kOutputRate = 8000;
const auto mixer =
AudioMixerImpl::Create(std::unique_ptr<OutputRateCalculator>(
new CustomRateCalculator(kOutputRate)),
true);
mixer->Mix(1, &frame_for_mixing);
EXPECT_EQ(kOutputRate, frame_for_mixing.sample_rate_hz_);
}
TEST(AudioMixer, NoLimiterBasicApiCalls) {
const auto mixer = AudioMixerImpl::Create(
std::unique_ptr<OutputRateCalculator>(new DefaultOutputRateCalculator()),
false);
mixer->Mix(1, &frame_for_mixing);
}
TEST(AudioMixer, AnyRateIsPossibleWithNoLimiter) {
// No APM limiter means no AudioProcessing::NativeRate restriction
// on mixing rate. The rate has to be divisible by 100 since we use
// 10 ms frames, though.
for (const auto rate : {8000, 20000, 24000, 32000, 44100}) {
for (const size_t number_of_channels : {1, 2}) {
for (const auto number_of_sources : {0, 1, 2, 3, 4}) {
SCOPED_TRACE(
ProduceDebugText(rate, number_of_sources, number_of_sources));
const auto mixer =
AudioMixerImpl::Create(std::unique_ptr<OutputRateCalculator>(
new CustomRateCalculator(rate)),
false);
std::vector<MockMixerAudioSource> sources(number_of_sources);
for (auto& source : sources) {
ResetFrame(source.fake_frame());
mixer->AddSource(&source);
}
mixer->Mix(number_of_channels, &frame_for_mixing);
EXPECT_EQ(rate, frame_for_mixing.sample_rate_hz_);
EXPECT_EQ(number_of_channels, frame_for_mixing.num_channels_);
}
}
}
}
TEST(AudioMixer, MultipleChannelsOneParticipant) {
// Set up a participant with a 6-channel frame, and make sure a 6-channel
// frame with the right sample values comes out from the mixer. There are 2
// Mix calls because of ramp-up.
constexpr size_t kNumberOfChannels = 6;
MockMixerAudioSource source;
ResetFrame(source.fake_frame());
const auto mixer = AudioMixerImpl::Create();
mixer->AddSource(&source);
mixer->Mix(1, &frame_for_mixing);
auto* frame = source.fake_frame();
frame->num_channels_ = kNumberOfChannels;
std::fill(frame->mutable_data(),
frame->mutable_data() + AudioFrame::kMaxDataSizeSamples, 0);
for (size_t i = 0; i < kNumberOfChannels; ++i) {
frame->mutable_data()[100 * frame->num_channels_ + i] = 1000 * i;
}
mixer->Mix(kNumberOfChannels, &frame_for_mixing);
EXPECT_EQ(frame_for_mixing.num_channels_, kNumberOfChannels);
for (size_t i = 0; i < kNumberOfChannels; ++i) {
EXPECT_EQ(frame_for_mixing.data()[100 * frame_for_mixing.num_channels_ + i],
static_cast<int16_t>(1000 * i));
}
}
TEST(AudioMixer, MultipleChannelsManyParticipants) {
// Sets up 2 participants. One has a 6-channel frame. Make sure a 6-channel
// frame with the right sample values comes out from the mixer. There are 2
// Mix calls because of ramp-up.
constexpr size_t kNumberOfChannels = 6;
MockMixerAudioSource source;
const auto mixer = AudioMixerImpl::Create();
mixer->AddSource(&source);
ResetFrame(source.fake_frame());
mixer->Mix(1, &frame_for_mixing);
auto* frame = source.fake_frame();
frame->num_channels_ = kNumberOfChannels;
std::fill(frame->mutable_data(),
frame->mutable_data() + AudioFrame::kMaxDataSizeSamples, 0);
for (size_t i = 0; i < kNumberOfChannels; ++i) {
frame->mutable_data()[100 * frame->num_channels_ + i] = 1000 * i;
}
MockMixerAudioSource other_source;
ResetFrame(other_source.fake_frame());
mixer->AddSource(&other_source);
mixer->Mix(kNumberOfChannels, &frame_for_mixing);
EXPECT_EQ(frame_for_mixing.num_channels_, kNumberOfChannels);
for (size_t i = 0; i < kNumberOfChannels; ++i) {
EXPECT_EQ(frame_for_mixing.data()[100 * frame_for_mixing.num_channels_ + i],
static_cast<int16_t>(1000 * i));
}
}
TEST(AudioMixer, ShouldIncludeRtpPacketInfoFromAllMixedSources) {
const uint32_t kSsrc0 = 10;
const uint32_t kSsrc1 = 11;
const uint32_t kSsrc2 = 12;
const uint32_t kCsrc0 = 20;
const uint32_t kCsrc1 = 21;
const uint32_t kCsrc2 = 22;
const uint32_t kCsrc3 = 23;
const int kAudioLevel0 = 10;
const int kAudioLevel1 = 40;
const absl::optional<uint32_t> kAudioLevel2 = absl::nullopt;
const uint32_t kRtpTimestamp0 = 300;
const uint32_t kRtpTimestamp1 = 400;
const Timestamp kReceiveTime0 = Timestamp::Millis(10);
const Timestamp kReceiveTime1 = Timestamp::Millis(20);
const RtpPacketInfo kPacketInfo0(kSsrc0, {kCsrc0, kCsrc1}, kRtpTimestamp0,
kAudioLevel0, absl::nullopt, kReceiveTime0);
const RtpPacketInfo kPacketInfo1(kSsrc1, {kCsrc2}, kRtpTimestamp1,
kAudioLevel1, absl::nullopt, kReceiveTime1);
const RtpPacketInfo kPacketInfo2(kSsrc2, {kCsrc3}, kRtpTimestamp1,
kAudioLevel2, absl::nullopt, kReceiveTime1);
const auto mixer = AudioMixerImpl::Create();
MockMixerAudioSource source;
source.set_packet_infos(RtpPacketInfos({kPacketInfo0}));
mixer->AddSource(&source);
ResetFrame(source.fake_frame());
mixer->Mix(1, &frame_for_mixing);
MockMixerAudioSource other_source;
other_source.set_packet_infos(RtpPacketInfos({kPacketInfo1, kPacketInfo2}));
ResetFrame(other_source.fake_frame());
mixer->AddSource(&other_source);
mixer->Mix(/*number_of_channels=*/1, &frame_for_mixing);
EXPECT_THAT(frame_for_mixing.packet_infos_,
UnorderedElementsAre(kPacketInfo0, kPacketInfo1, kPacketInfo2));
}
TEST(AudioMixer, MixerShouldIncludeRtpPacketInfoFromMixedSourcesOnly) {
const uint32_t kSsrc0 = 10;
const uint32_t kSsrc1 = 11;
const uint32_t kSsrc2 = 21;
const uint32_t kCsrc0 = 30;
const uint32_t kCsrc1 = 31;
const uint32_t kCsrc2 = 32;
const uint32_t kCsrc3 = 33;
const int kAudioLevel0 = 10;
const absl::optional<uint32_t> kAudioLevelMissing = absl::nullopt;
const uint32_t kRtpTimestamp0 = 300;
const uint32_t kRtpTimestamp1 = 400;
const Timestamp kReceiveTime0 = Timestamp::Millis(10);
const Timestamp kReceiveTime1 = Timestamp::Millis(20);
const RtpPacketInfo kPacketInfo0(kSsrc0, {kCsrc0, kCsrc1}, kRtpTimestamp0,
kAudioLevel0, absl::nullopt, kReceiveTime0);
const RtpPacketInfo kPacketInfo1(kSsrc1, {kCsrc2}, kRtpTimestamp1,
kAudioLevelMissing, absl::nullopt,
kReceiveTime1);
const RtpPacketInfo kPacketInfo2(kSsrc2, {kCsrc3}, kRtpTimestamp1,
kAudioLevelMissing, absl::nullopt,
kReceiveTime1);
const auto mixer = AudioMixerImpl::Create(/*max_sources_to_mix=*/2);
MockMixerAudioSource source1;
source1.set_packet_infos(RtpPacketInfos({kPacketInfo0}));
mixer->AddSource(&source1);
ResetFrame(source1.fake_frame());
mixer->Mix(1, &frame_for_mixing);
MockMixerAudioSource source2;
source2.set_packet_infos(RtpPacketInfos({kPacketInfo1}));
ResetFrame(source2.fake_frame());
mixer->AddSource(&source2);
// The mixer prioritizes kVadActive over kVadPassive.
// We limit the number of sources to mix to 2 and set the third source's VAD
// activity to kVadPassive so that it will not be added to the mix.
MockMixerAudioSource source3;
source3.set_packet_infos(RtpPacketInfos({kPacketInfo2}));
ResetFrame(source3.fake_frame());
source3.fake_frame()->vad_activity_ = AudioFrame::kVadPassive;
mixer->AddSource(&source3);
mixer->Mix(/*number_of_channels=*/1, &frame_for_mixing);
EXPECT_THAT(frame_for_mixing.packet_infos_,
UnorderedElementsAre(kPacketInfo0, kPacketInfo1));
}
class HighOutputRateCalculator : public OutputRateCalculator {
public:
static const int kDefaultFrequency = 76000;
int CalculateOutputRateFromRange(
rtc::ArrayView<const int> preferred_sample_rates) override {
return kDefaultFrequency;
}
~HighOutputRateCalculator() override {}
};
const int HighOutputRateCalculator::kDefaultFrequency;
TEST(AudioMixerDeathTest, MultipleChannelsAndHighRate) {
constexpr size_t kSamplesPerChannel =
HighOutputRateCalculator::kDefaultFrequency / 100;
// As many channels as an AudioFrame can fit:
constexpr size_t kNumberOfChannels =
AudioFrame::kMaxDataSizeSamples / kSamplesPerChannel;
MockMixerAudioSource source;
const auto mixer = AudioMixerImpl::Create(
std::make_unique<HighOutputRateCalculator>(), true);
mixer->AddSource(&source);
ResetFrame(source.fake_frame());
mixer->Mix(1, &frame_for_mixing);
auto* frame = source.fake_frame();
frame->num_channels_ = kNumberOfChannels;
frame->sample_rate_hz_ = HighOutputRateCalculator::kDefaultFrequency;
frame->samples_per_channel_ = kSamplesPerChannel;
std::fill(frame->mutable_data(),
frame->mutable_data() + AudioFrame::kMaxDataSizeSamples, 0);
MockMixerAudioSource other_source;
ResetFrame(other_source.fake_frame());
auto* other_frame = other_source.fake_frame();
other_frame->num_channels_ = kNumberOfChannels;
other_frame->sample_rate_hz_ = HighOutputRateCalculator::kDefaultFrequency;
other_frame->samples_per_channel_ = kSamplesPerChannel;
mixer->AddSource(&other_source);
#if RTC_DCHECK_IS_ON && GTEST_HAS_DEATH_TEST && !defined(WEBRTC_ANDROID)
EXPECT_DEATH(mixer->Mix(kNumberOfChannels, &frame_for_mixing), "");
#elif !RTC_DCHECK_IS_ON
mixer->Mix(kNumberOfChannels, &frame_for_mixing);
EXPECT_EQ(frame_for_mixing.num_channels_, kNumberOfChannels);
EXPECT_EQ(frame_for_mixing.sample_rate_hz_,
HighOutputRateCalculator::kDefaultFrequency);
#endif
}
} // namespace webrtc