blob: d3af7c03418851a16c49bfc3254cacb2ab3d0bd7 [file] [log] [blame]
/*
* Copyright (c) 2013 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_coding/acm2/acm_receiver.h"
#include <stdlib.h>
#include <string.h>
#include <cstdint>
#include <vector>
#include "absl/strings/match.h"
#include "api/audio/audio_frame.h"
#include "api/audio_codecs/audio_decoder.h"
#include "modules/audio_coding/acm2/acm_resampler.h"
#include "modules/audio_coding/acm2/call_statistics.h"
#include "modules/audio_coding/acm2/rent_a_codec.h"
#include "modules/audio_coding/neteq/include/neteq.h"
#include "modules/audio_coding/neteq/neteq_decoder_enum.h"
#include "modules/include/module_common_types.h"
#include "rtc_base/checks.h"
#include "rtc_base/logging.h"
#include "rtc_base/numerics/safe_conversions.h"
#include "rtc_base/strings/audio_format_to_string.h"
#include "system_wrappers/include/clock.h"
namespace webrtc {
namespace acm2 {
AcmReceiver::AcmReceiver(const AudioCodingModule::Config& config)
: last_audio_buffer_(new int16_t[AudioFrame::kMaxDataSizeSamples]),
neteq_(NetEq::Create(config.neteq_config, config.decoder_factory)),
clock_(config.clock),
resampled_last_output_frame_(true) {
RTC_DCHECK(clock_);
memset(last_audio_buffer_.get(), 0,
sizeof(int16_t) * AudioFrame::kMaxDataSizeSamples);
}
AcmReceiver::~AcmReceiver() = default;
int AcmReceiver::SetMinimumDelay(int delay_ms) {
if (neteq_->SetMinimumDelay(delay_ms))
return 0;
RTC_LOG(LERROR) << "AcmReceiver::SetExtraDelay " << delay_ms;
return -1;
}
int AcmReceiver::SetMaximumDelay(int delay_ms) {
if (neteq_->SetMaximumDelay(delay_ms))
return 0;
RTC_LOG(LERROR) << "AcmReceiver::SetExtraDelay " << delay_ms;
return -1;
}
absl::optional<int> AcmReceiver::last_packet_sample_rate_hz() const {
rtc::CritScope lock(&crit_sect_);
return last_packet_sample_rate_hz_;
}
int AcmReceiver::last_output_sample_rate_hz() const {
return neteq_->last_output_sample_rate_hz();
}
int AcmReceiver::InsertPacket(const WebRtcRTPHeader& rtp_header,
rtc::ArrayView<const uint8_t> incoming_payload) {
uint32_t receive_timestamp = 0;
const RTPHeader* header = &rtp_header.header; // Just a shorthand.
if (incoming_payload.empty()) {
neteq_->InsertEmptyPacket(rtp_header.header);
return 0;
}
{
rtc::CritScope lock(&crit_sect_);
const absl::optional<CodecInst> ci =
RtpHeaderToDecoder(*header, incoming_payload[0]);
if (!ci) {
RTC_LOG_F(LS_ERROR) << "Payload-type "
<< static_cast<int>(header->payloadType)
<< " is not registered.";
return -1;
}
receive_timestamp = NowInTimestamp(ci->plfreq);
if (absl::EqualsIgnoreCase(ci->plname, "cn")) {
if (last_audio_decoder_ && last_audio_decoder_->channels > 1) {
// This is a CNG and the audio codec is not mono, so skip pushing in
// packets into NetEq.
return 0;
}
} else {
last_audio_decoder_ = ci;
last_audio_format_ = neteq_->GetDecoderFormat(ci->pltype);
RTC_DCHECK(last_audio_format_);
last_packet_sample_rate_hz_ = ci->plfreq;
}
} // |crit_sect_| is released.
if (neteq_->InsertPacket(rtp_header.header, incoming_payload,
receive_timestamp) < 0) {
RTC_LOG(LERROR) << "AcmReceiver::InsertPacket "
<< static_cast<int>(header->payloadType)
<< " Failed to insert packet";
return -1;
}
return 0;
}
int AcmReceiver::GetAudio(int desired_freq_hz,
AudioFrame* audio_frame,
bool* muted) {
RTC_DCHECK(muted);
// Accessing members, take the lock.
rtc::CritScope lock(&crit_sect_);
if (neteq_->GetAudio(audio_frame, muted) != NetEq::kOK) {
RTC_LOG(LERROR) << "AcmReceiver::GetAudio - NetEq Failed.";
return -1;
}
const int current_sample_rate_hz = neteq_->last_output_sample_rate_hz();
// Update if resampling is required.
const bool need_resampling =
(desired_freq_hz != -1) && (current_sample_rate_hz != desired_freq_hz);
if (need_resampling && !resampled_last_output_frame_) {
// Prime the resampler with the last frame.
int16_t temp_output[AudioFrame::kMaxDataSizeSamples];
int samples_per_channel_int = resampler_.Resample10Msec(
last_audio_buffer_.get(), current_sample_rate_hz, desired_freq_hz,
audio_frame->num_channels_, AudioFrame::kMaxDataSizeSamples,
temp_output);
if (samples_per_channel_int < 0) {
RTC_LOG(LERROR) << "AcmReceiver::GetAudio - "
"Resampling last_audio_buffer_ failed.";
return -1;
}
}
// TODO(henrik.lundin) Glitches in the output may appear if the output rate
// from NetEq changes. See WebRTC issue 3923.
if (need_resampling) {
// TODO(yujo): handle this more efficiently for muted frames.
int samples_per_channel_int = resampler_.Resample10Msec(
audio_frame->data(), current_sample_rate_hz, desired_freq_hz,
audio_frame->num_channels_, AudioFrame::kMaxDataSizeSamples,
audio_frame->mutable_data());
if (samples_per_channel_int < 0) {
RTC_LOG(LERROR)
<< "AcmReceiver::GetAudio - Resampling audio_buffer_ failed.";
return -1;
}
audio_frame->samples_per_channel_ =
static_cast<size_t>(samples_per_channel_int);
audio_frame->sample_rate_hz_ = desired_freq_hz;
RTC_DCHECK_EQ(
audio_frame->sample_rate_hz_,
rtc::dchecked_cast<int>(audio_frame->samples_per_channel_ * 100));
resampled_last_output_frame_ = true;
} else {
resampled_last_output_frame_ = false;
// We might end up here ONLY if codec is changed.
}
// Store current audio in |last_audio_buffer_| for next time.
memcpy(last_audio_buffer_.get(), audio_frame->data(),
sizeof(int16_t) * audio_frame->samples_per_channel_ *
audio_frame->num_channels_);
call_stats_.DecodedByNetEq(audio_frame->speech_type_, *muted);
return 0;
}
void AcmReceiver::SetCodecs(const std::map<int, SdpAudioFormat>& codecs) {
neteq_->SetCodecs(codecs);
}
int32_t AcmReceiver::AddCodec(int acm_codec_id,
uint8_t payload_type,
size_t channels,
int /*sample_rate_hz*/,
AudioDecoder* audio_decoder,
const std::string& name) {
// TODO(kwiberg): This function has been ignoring the |sample_rate_hz|
// argument for a long time. Arguably, it should simply be removed.
const auto neteq_decoder = [acm_codec_id, channels]() -> NetEqDecoder {
if (acm_codec_id == -1)
return NetEqDecoder::kDecoderArbitrary; // External decoder.
const absl::optional<RentACodec::CodecId> cid =
RentACodec::CodecIdFromIndex(acm_codec_id);
RTC_DCHECK(cid) << "Invalid codec index: " << acm_codec_id;
const absl::optional<NetEqDecoder> ned =
RentACodec::NetEqDecoderFromCodecId(*cid, channels);
RTC_DCHECK(ned) << "Invalid codec ID: " << static_cast<int>(*cid);
return *ned;
}();
const absl::optional<SdpAudioFormat> new_format =
NetEqDecoderToSdpAudioFormat(neteq_decoder);
rtc::CritScope lock(&crit_sect_);
const auto old_format = neteq_->GetDecoderFormat(payload_type);
if (old_format && new_format && *old_format == *new_format) {
// Re-registering the same codec. Do nothing and return.
return 0;
}
if (neteq_->RemovePayloadType(payload_type) != NetEq::kOK) {
RTC_LOG(LERROR) << "Cannot remove payload "
<< static_cast<int>(payload_type);
return -1;
}
int ret_val;
if (!audio_decoder) {
ret_val = neteq_->RegisterPayloadType(neteq_decoder, name, payload_type);
} else {
ret_val = neteq_->RegisterExternalDecoder(audio_decoder, neteq_decoder,
name, payload_type);
}
if (ret_val != NetEq::kOK) {
RTC_LOG(LERROR) << "AcmReceiver::AddCodec " << acm_codec_id
<< static_cast<int>(payload_type)
<< " channels: " << channels;
return -1;
}
return 0;
}
bool AcmReceiver::AddCodec(int rtp_payload_type,
const SdpAudioFormat& audio_format) {
const auto old_format = neteq_->GetDecoderFormat(rtp_payload_type);
if (old_format && *old_format == audio_format) {
// Re-registering the same codec. Do nothing and return.
return true;
}
if (neteq_->RemovePayloadType(rtp_payload_type) != NetEq::kOK) {
RTC_LOG(LERROR)
<< "AcmReceiver::AddCodec: Could not remove existing decoder"
" for payload type "
<< rtp_payload_type;
return false;
}
const bool success =
neteq_->RegisterPayloadType(rtp_payload_type, audio_format);
if (!success) {
RTC_LOG(LERROR) << "AcmReceiver::AddCodec failed for payload type "
<< rtp_payload_type << ", decoder format "
<< rtc::ToString(audio_format);
}
return success;
}
void AcmReceiver::FlushBuffers() {
neteq_->FlushBuffers();
}
void AcmReceiver::RemoveAllCodecs() {
rtc::CritScope lock(&crit_sect_);
neteq_->RemoveAllPayloadTypes();
last_audio_decoder_ = absl::nullopt;
last_audio_format_ = absl::nullopt;
last_packet_sample_rate_hz_ = absl::nullopt;
}
int AcmReceiver::RemoveCodec(uint8_t payload_type) {
rtc::CritScope lock(&crit_sect_);
if (neteq_->RemovePayloadType(payload_type) != NetEq::kOK) {
RTC_LOG(LERROR) << "AcmReceiver::RemoveCodec "
<< static_cast<int>(payload_type);
return -1;
}
if (last_audio_decoder_ && payload_type == last_audio_decoder_->pltype) {
last_audio_decoder_ = absl::nullopt;
last_audio_format_ = absl::nullopt;
last_packet_sample_rate_hz_ = absl::nullopt;
}
return 0;
}
absl::optional<uint32_t> AcmReceiver::GetPlayoutTimestamp() {
return neteq_->GetPlayoutTimestamp();
}
int AcmReceiver::FilteredCurrentDelayMs() const {
return neteq_->FilteredCurrentDelayMs();
}
int AcmReceiver::TargetDelayMs() const {
return neteq_->TargetDelayMs();
}
int AcmReceiver::LastAudioCodec(CodecInst* codec) const {
rtc::CritScope lock(&crit_sect_);
if (!last_audio_decoder_) {
return -1;
}
*codec = *last_audio_decoder_;
return 0;
}
absl::optional<SdpAudioFormat> AcmReceiver::LastAudioFormat() const {
rtc::CritScope lock(&crit_sect_);
return last_audio_format_;
}
void AcmReceiver::GetNetworkStatistics(NetworkStatistics* acm_stat) {
NetEqNetworkStatistics neteq_stat;
// NetEq function always returns zero, so we don't check the return value.
neteq_->NetworkStatistics(&neteq_stat);
acm_stat->currentBufferSize = neteq_stat.current_buffer_size_ms;
acm_stat->preferredBufferSize = neteq_stat.preferred_buffer_size_ms;
acm_stat->jitterPeaksFound = neteq_stat.jitter_peaks_found ? true : false;
acm_stat->currentPacketLossRate = neteq_stat.packet_loss_rate;
acm_stat->currentExpandRate = neteq_stat.expand_rate;
acm_stat->currentSpeechExpandRate = neteq_stat.speech_expand_rate;
acm_stat->currentPreemptiveRate = neteq_stat.preemptive_rate;
acm_stat->currentAccelerateRate = neteq_stat.accelerate_rate;
acm_stat->currentSecondaryDecodedRate = neteq_stat.secondary_decoded_rate;
acm_stat->currentSecondaryDiscardedRate = neteq_stat.secondary_discarded_rate;
acm_stat->clockDriftPPM = neteq_stat.clockdrift_ppm;
acm_stat->addedSamples = neteq_stat.added_zero_samples;
acm_stat->meanWaitingTimeMs = neteq_stat.mean_waiting_time_ms;
acm_stat->medianWaitingTimeMs = neteq_stat.median_waiting_time_ms;
acm_stat->minWaitingTimeMs = neteq_stat.min_waiting_time_ms;
acm_stat->maxWaitingTimeMs = neteq_stat.max_waiting_time_ms;
NetEqLifetimeStatistics neteq_lifetime_stat = neteq_->GetLifetimeStatistics();
acm_stat->totalSamplesReceived = neteq_lifetime_stat.total_samples_received;
acm_stat->concealedSamples = neteq_lifetime_stat.concealed_samples;
acm_stat->concealmentEvents = neteq_lifetime_stat.concealment_events;
acm_stat->jitterBufferDelayMs = neteq_lifetime_stat.jitter_buffer_delay_ms;
acm_stat->delayedPacketOutageSamples =
neteq_lifetime_stat.delayed_packet_outage_samples;
NetEqOperationsAndState neteq_operations_and_state =
neteq_->GetOperationsAndState();
acm_stat->packetBufferFlushes =
neteq_operations_and_state.packet_buffer_flushes;
}
int AcmReceiver::DecoderByPayloadType(uint8_t payload_type,
CodecInst* codec) const {
rtc::CritScope lock(&crit_sect_);
const absl::optional<CodecInst> ci = neteq_->GetDecoder(payload_type);
if (ci) {
*codec = *ci;
return 0;
} else {
RTC_LOG(LERROR) << "AcmReceiver::DecoderByPayloadType "
<< static_cast<int>(payload_type);
return -1;
}
}
absl::optional<SdpAudioFormat> AcmReceiver::DecoderByPayloadType(
int payload_type) const {
rtc::CritScope lock(&crit_sect_);
return neteq_->GetDecoderFormat(payload_type);
}
int AcmReceiver::EnableNack(size_t max_nack_list_size) {
neteq_->EnableNack(max_nack_list_size);
return 0;
}
void AcmReceiver::DisableNack() {
neteq_->DisableNack();
}
std::vector<uint16_t> AcmReceiver::GetNackList(
int64_t round_trip_time_ms) const {
return neteq_->GetNackList(round_trip_time_ms);
}
void AcmReceiver::ResetInitialDelay() {
neteq_->SetMinimumDelay(0);
// TODO(turajs): Should NetEq Buffer be flushed?
}
const absl::optional<CodecInst> AcmReceiver::RtpHeaderToDecoder(
const RTPHeader& rtp_header,
uint8_t first_payload_byte) const {
const absl::optional<CodecInst> ci =
neteq_->GetDecoder(rtp_header.payloadType);
if (ci && absl::EqualsIgnoreCase(ci->plname, "red")) {
// This is a RED packet. Get the payload of the audio codec.
return neteq_->GetDecoder(first_payload_byte & 0x7f);
} else {
return ci;
}
}
uint32_t AcmReceiver::NowInTimestamp(int decoder_sampling_rate) const {
// Down-cast the time to (32-6)-bit since we only care about
// the least significant bits. (32-6) bits cover 2^(32-6) = 67108864 ms.
// We masked 6 most significant bits of 32-bit so there is no overflow in
// the conversion from milliseconds to timestamp.
const uint32_t now_in_ms =
static_cast<uint32_t>(clock_->TimeInMilliseconds() & 0x03ffffff);
return static_cast<uint32_t>((decoder_sampling_rate / 1000) * now_in_ms);
}
void AcmReceiver::GetDecodingCallStatistics(
AudioDecodingCallStats* stats) const {
rtc::CritScope lock(&crit_sect_);
*stats = call_stats_.GetDecodingStatistics();
}
} // namespace acm2
} // namespace webrtc