-/*\r
-* copyright (c) 2010 Sveriges Television AB <info@casparcg.com>\r
-*\r
-* This file is part of CasparCG.\r
-*\r
-* CasparCG is free software: you can redistribute it and/or modify\r
-* it under the terms of the GNU General Public License as published by\r
-* the Free Software Foundation, either version 3 of the License, or\r
-* (at your option) any later version.\r
-*\r
-* CasparCG is distributed in the hope that it will be useful,\r
-* but WITHOUT ANY WARRANTY; without even the implied warranty of\r
-* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\r
-* GNU General Public License for more details.\r
-\r
-* You should have received a copy of the GNU General Public License\r
-* along with CasparCG. If not, see <http://www.gnu.org/licenses/>.\r
-*\r
-*/\r
-#include "../../stdafx.h"\r
-\r
-#include "audio_mixer.h"\r
-\r
-#include <core/mixer/write_frame.h>\r
-#include <core/producer/frame/audio_transform.h>\r
-\r
-namespace caspar { namespace core {\r
- \r
-struct audio_mixer::implementation\r
-{\r
- std::deque<std::vector<int16_t>> audio_data_;\r
- std::stack<core::audio_transform> transform_stack_;\r
-\r
- std::map<int, core::audio_transform> prev_audio_transforms_;\r
- std::map<int, core::audio_transform> next_audio_transforms_;\r
-\r
-public:\r
- implementation()\r
- {\r
- transform_stack_.push(core::audio_transform());\r
- audio_data_.push_back(std::vector<int16_t>()); // One frame delay\r
- }\r
- \r
- void begin(const core::basic_frame& frame)\r
- {\r
- transform_stack_.push(transform_stack_.top()*frame.get_audio_transform());\r
- }\r
-\r
- void visit(const core::write_frame& frame)\r
- {\r
- if(!transform_stack_.top().get_has_audio())\r
- return;\r
-\r
- auto& audio_data = frame.audio_data();\r
- auto tag = frame.tag(); // Get the identifier for the audio-stream.\r
-\r
- if(audio_data_.back().empty())\r
- audio_data_.back().resize(audio_data.size(), 0);\r
- \r
- auto next = transform_stack_.top();\r
- auto prev = next;\r
-\r
- auto it = prev_audio_transforms_.find(tag);\r
- if(it != prev_audio_transforms_.end())\r
- prev = it->second;\r
- \r
- next_audio_transforms_[tag] = next; // Store all active tags, inactive tags will be removed in end_pass.\r
- \r
- if(next.get_gain() < 0.001 && prev.get_gain() < 0.001)\r
- return;\r
- \r
- static const int BASE = 1<<15;\r
-\r
- auto next_gain = static_cast<int>(next.get_gain()*BASE);\r
- auto prev_gain = static_cast<int>(prev.get_gain()*BASE);\r
- \r
- int n_samples = audio_data_.back().size();\r
-\r
- tbb::parallel_for\r
- (\r
- tbb::blocked_range<size_t>(0, audio_data.size()),\r
- [&](const tbb::blocked_range<size_t>& r)\r
- {\r
- for(size_t n = r.begin(); n < r.end(); ++n)\r
- {\r
- int sample_gain = (prev_gain - (prev_gain * n)/n_samples) + (next_gain * n)/n_samples;\r
- \r
- int sample = (static_cast<int>(audio_data[n])*sample_gain)/BASE;\r
- \r
- audio_data_.back()[n] = static_cast<int16_t>((static_cast<int>(audio_data_.back()[n]) + sample) & 0xFFFF);\r
- }\r
- }\r
- );\r
- }\r
-\r
- void begin(const core::audio_transform& transform)\r
- {\r
- transform_stack_.push(transform_stack_.top()*transform);\r
- }\r
- \r
- void end()\r
- {\r
- transform_stack_.pop();\r
- }\r
- \r
- std::vector<int16_t> mix()\r
- {\r
- prev_audio_transforms_ = std::move(next_audio_transforms_); \r
- auto result = std::move(audio_data_.front());\r
- audio_data_.pop_front();\r
- audio_data_.push_back(std::vector<int16_t>());\r
- return std::move(result);\r
- }\r
-};\r
-\r
-audio_mixer::audio_mixer() : impl_(new implementation()){}\r
-void audio_mixer::begin(const core::basic_frame& frame){impl_->begin(frame);}\r
-void audio_mixer::visit(core::write_frame& frame){impl_->visit(frame);}\r
-void audio_mixer::end(){impl_->end();}\r
-std::vector<int16_t> audio_mixer::mix(){return impl_->mix();}\r
-audio_mixer& audio_mixer::operator=(audio_mixer&& other)\r
-{\r
- impl_ = std::move(other.impl_);\r
- return *this;\r
-}\r
-\r
-}}
\ No newline at end of file
+/*
+* Copyright (c) 2011 Sveriges Television AB <info@casparcg.com>
+*
+* This file is part of CasparCG (www.casparcg.com).
+*
+* CasparCG is free software: you can redistribute it and/or modify
+* it under the terms of the GNU General Public License as published by
+* the Free Software Foundation, either version 3 of the License, or
+* (at your option) any later version.
+*
+* CasparCG is distributed in the hope that it will be useful,
+* but WITHOUT ANY WARRANTY; without even the implied warranty of
+* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+* GNU General Public License for more details.
+*
+* You should have received a copy of the GNU General Public License
+* along with CasparCG. If not, see <http://www.gnu.org/licenses/>.
+*
+* Author: Robert Nagy, ronag89@gmail.com
+*/
+
+#include "../../StdAfx.h"
+
+#include "audio_mixer.h"
+
+#include <core/frame/frame.h>
+#include <core/frame/frame_transform.h>
+#include <core/frame/audio_channel_layout.h>
+#include <core/monitor/monitor.h>
+
+#include <common/diagnostics/graph.h>
+#include <common/linq.h>
+
+#include <boost/range/adaptors.hpp>
+#include <boost/range/distance.hpp>
+#include <boost/lexical_cast.hpp>
+
+#include <map>
+#include <stack>
+#include <vector>
+
+namespace caspar { namespace core {
+
+struct audio_item
+{
+ const void* tag = nullptr;
+ audio_transform transform;
+ audio_buffer audio_data;
+ audio_channel_layout channel_layout = audio_channel_layout::invalid();
+
+ audio_item()
+ {
+ }
+
+ audio_item(audio_item&& other)
+ : tag(std::move(other.tag))
+ , transform(std::move(other.transform))
+ , audio_data(std::move(other.audio_data))
+ , channel_layout(std::move(other.channel_layout))
+ {
+ }
+};
+
+typedef cache_aligned_vector<double> audio_buffer_ps;
+
+struct audio_stream
+{
+ audio_transform prev_transform;
+ audio_buffer_ps audio_data;
+ std::unique_ptr<audio_channel_remapper> channel_remapper;
+ bool remapping_failed = false;
+ bool is_still = false;
+};
+
+struct audio_mixer::impl : boost::noncopyable
+{
+ monitor::subject monitor_subject_ { "/audio" };
+ std::stack<core::audio_transform> transform_stack_;
+ std::map<const void*, audio_stream> audio_streams_;
+ std::vector<audio_item> items_;
+ std::vector<int> audio_cadence_;
+ video_format_desc format_desc_;
+ audio_channel_layout channel_layout_ = audio_channel_layout::invalid();
+ float master_volume_ = 1.0f;
+ float previous_master_volume_ = master_volume_;
+ spl::shared_ptr<diagnostics::graph> graph_;
+public:
+ impl(spl::shared_ptr<diagnostics::graph> graph)
+ : graph_(std::move(graph))
+ {
+ graph_->set_color("volume", diagnostics::color(1.0f, 0.8f, 0.1f));
+ graph_->set_color("audio-clipping", diagnostics::color(0.3f, 0.6f, 0.3f));
+ transform_stack_.push(core::audio_transform());
+ }
+
+ void push(const frame_transform& transform)
+ {
+ transform_stack_.push(transform_stack_.top()*transform.audio_transform);
+ }
+
+ void visit(const const_frame& frame)
+ {
+ if(transform_stack_.top().volume < 0.002 || frame.audio_data().empty())
+ return;
+
+ audio_item item;
+ item.tag = frame.stream_tag();
+ item.transform = transform_stack_.top();
+ item.audio_data = frame.audio_data();
+ item.channel_layout = frame.audio_channel_layout();
+
+ if(item.transform.is_still)
+ item.transform.volume = 0.0;
+
+ items_.push_back(std::move(item));
+ }
+
+ void pop()
+ {
+ transform_stack_.pop();
+ }
+
+ void set_master_volume(float volume)
+ {
+ master_volume_ = volume;
+ }
+
+ float get_master_volume()
+ {
+ return master_volume_;
+ }
+
+ audio_buffer mix(const video_format_desc& format_desc, const audio_channel_layout& channel_layout)
+ {
+ if(format_desc_ != format_desc || channel_layout_ != channel_layout)
+ {
+ audio_streams_.clear();
+ audio_cadence_ = format_desc.audio_cadence;
+ format_desc_ = format_desc;
+ channel_layout_ = channel_layout;
+ }
+
+ std::map<const void*, audio_stream> next_audio_streams;
+
+ for (auto& item : items_)
+ {
+ audio_buffer_ps next_audio;
+ std::unique_ptr<audio_channel_remapper> channel_remapper;
+ bool remapping_failed = false;
+
+ auto next_transform = item.transform;
+ auto prev_transform = next_transform;
+
+ auto tag = item.tag;
+
+ auto it = next_audio_streams.find(tag);
+ bool found = it != next_audio_streams.end();
+
+ if (!found)
+ {
+ it = audio_streams_.find(tag);
+ found = it != audio_streams_.end();
+ }
+
+ if (found)
+ {
+ prev_transform = it->second.prev_transform;
+ next_audio = std::move(it->second.audio_data);
+ channel_remapper = std::move(it->second.channel_remapper);
+ remapping_failed = it->second.remapping_failed;
+ }
+
+ if (remapping_failed)
+ {
+ CASPAR_LOG(trace) << "[audio_mixer] audio channel remapping already failed for stream.";
+ next_audio_streams[tag].remapping_failed = true;
+ continue;
+ }
+
+ // Skip it if there is no existing audio stream and item has no audio-data.
+ if(!found && item.audio_data.empty())
+ continue;
+
+ if (item.channel_layout == audio_channel_layout::invalid())
+ {
+ CASPAR_LOG(warning) << "[audio_mixer] invalid audio channel layout for item";
+ next_audio_streams[tag].remapping_failed = true;
+ continue;
+ }
+
+ if (!channel_remapper)
+ {
+ try
+ {
+ channel_remapper.reset(new audio_channel_remapper(item.channel_layout, channel_layout_));
+ }
+ catch (...)
+ {
+ CASPAR_LOG_CURRENT_EXCEPTION();
+ CASPAR_LOG(error) << "[audio_mixer] audio channel remapping failed for stream.";
+ next_audio_streams[tag].remapping_failed = true;
+ continue;
+ }
+ }
+
+ item.audio_data = channel_remapper->mix_and_rearrange(item.audio_data);
+
+ const double prev_volume = prev_transform.volume * previous_master_volume_;
+ const double next_volume = next_transform.volume * master_volume_;
+
+ // TODO: Move volume mixing into code below, in order to support audio sample counts not corresponding to frame audio samples.
+ auto alpha = (next_volume-prev_volume)/static_cast<double>(item.audio_data.size()/channel_layout_.num_channels);
+
+ for(size_t n = 0; n < item.audio_data.size(); ++n)
+ {
+ auto sample_multiplier = (prev_volume + (n / channel_layout_.num_channels) * alpha);
+ next_audio.push_back(item.audio_data.data()[n] * sample_multiplier);
+ }
+
+ next_audio_streams[tag].prev_transform = std::move(next_transform); // Store all active tags, inactive tags will be removed at the end.
+ next_audio_streams[tag].audio_data = std::move(next_audio);
+ next_audio_streams[tag].channel_remapper = std::move(channel_remapper);
+ next_audio_streams[tag].remapping_failed = remapping_failed;
+ next_audio_streams[tag].is_still = item.transform.is_still;
+ }
+
+ previous_master_volume_ = master_volume_;
+ items_.clear();
+
+ audio_streams_ = std::move(next_audio_streams);
+
+ if(audio_streams_.empty())
+ audio_streams_[nullptr].audio_data = audio_buffer_ps(audio_size(audio_cadence_.front()), 0.0);
+
+ { // sanity check
+
+ auto nb_invalid_streams = cpplinq::from(audio_streams_)
+ .select(values())
+ .where([&](const audio_stream& x)
+ {
+ return !x.remapping_failed && x.audio_data.size() < audio_size(audio_cadence_.front());
+ })
+ .count();
+
+ if(nb_invalid_streams > 0)
+ CASPAR_LOG(trace) << "[audio_mixer] Incorrect frame audio cadence detected.";
+ }
+
+ audio_buffer_ps result_ps(audio_size(audio_cadence_.front()), 0.0);
+ for (auto& stream : audio_streams_ | boost::adaptors::map_values)
+ {
+ if (stream.audio_data.size() < result_ps.size())
+ {
+ auto samples = (result_ps.size() - stream.audio_data.size()) / channel_layout_.num_channels;
+ CASPAR_LOG(trace) << L"[audio_mixer] Appended " << samples << L" zero samples";
+ CASPAR_LOG(trace) << L"[audio_mixer] Actual number of samples " << stream.audio_data.size() / channel_layout_.num_channels;
+ CASPAR_LOG(trace) << L"[audio_mixer] Wanted number of samples " << result_ps.size() / channel_layout_.num_channels;
+ stream.audio_data.resize(result_ps.size(), 0.0);
+ }
+
+ auto out = boost::range::transform(result_ps, stream.audio_data, std::begin(result_ps), std::plus<double>());
+ stream.audio_data.erase(std::begin(stream.audio_data), std::begin(stream.audio_data) + std::distance(std::begin(result_ps), out));
+ }
+
+ boost::range::rotate(audio_cadence_, std::begin(audio_cadence_)+1);
+
+ auto result_owner = spl::make_shared<mutable_audio_buffer>();
+ auto& result = *result_owner;
+ result.reserve(result_ps.size());
+ const int32_t min_amplitude = std::numeric_limits<int32_t>::min();
+ const int32_t max_amplitude = std::numeric_limits<int32_t>::max();
+ bool clipping = false;
+ boost::range::transform(result_ps, std::back_inserter(result), [&](double sample)
+ {
+ if (sample > max_amplitude)
+ {
+ clipping = true;
+ return max_amplitude;
+ }
+ else if (sample < min_amplitude)
+ {
+ clipping = true;
+ return min_amplitude;
+ }
+ else
+ return static_cast<int32_t>(sample);
+ });
+
+ if (clipping)
+ graph_->set_tag(diagnostics::tag_severity::WARNING, "audio-clipping");
+
+ const int num_channels = channel_layout_.num_channels;
+ monitor_subject_ << monitor::message("/nb_channels") % num_channels;
+
+ auto max = std::vector<int32_t>(num_channels, std::numeric_limits<int32_t>::min());
+
+ for (size_t n = 0; n < result.size(); n += num_channels)
+ for (int ch = 0; ch < num_channels; ++ch)
+ max[ch] = std::max(max[ch], std::abs(result[n + ch]));
+
+ // Makes the dBFS of silence => -dynamic range of 32bit LPCM => about -192 dBFS
+ // Otherwise it would be -infinity
+ static const auto MIN_PFS = 0.5f / static_cast<float>(std::numeric_limits<int32_t>::max());
+
+ for (int i = 0; i < num_channels; ++i)
+ {
+ const auto pFS = max[i] / static_cast<float>(std::numeric_limits<int32_t>::max());
+ const auto dBFS = 20.0f * std::log10(std::max(MIN_PFS, pFS));
+
+ auto chan_str = boost::lexical_cast<std::string>(i + 1);
+
+ monitor_subject_ << monitor::message("/" + chan_str + "/pFS") % pFS;
+ monitor_subject_ << monitor::message("/" + chan_str + "/dBFS") % dBFS;
+ }
+
+ graph_->set_value("volume", static_cast<double>(*boost::max_element(max)) / std::numeric_limits<int32_t>::max());
+
+ return caspar::array<int32_t>(result.data(), result.size(), true, std::move(result_owner));
+ }
+
+ size_t audio_size(size_t num_samples) const
+ {
+ return num_samples * channel_layout_.num_channels;
+ }
+};
+
+audio_mixer::audio_mixer(spl::shared_ptr<diagnostics::graph> graph) : impl_(new impl(std::move(graph))){}
+void audio_mixer::push(const frame_transform& transform){impl_->push(transform);}
+void audio_mixer::visit(const const_frame& frame){impl_->visit(frame);}
+void audio_mixer::pop(){impl_->pop();}
+void audio_mixer::set_master_volume(float volume) { impl_->set_master_volume(volume); }
+float audio_mixer::get_master_volume() { return impl_->get_master_volume(); }
+audio_buffer audio_mixer::operator()(const video_format_desc& format_desc, const audio_channel_layout& channel_layout){ return impl_->mix(format_desc, channel_layout); }
+monitor::subject& audio_mixer::monitor_output(){ return impl_->monitor_subject_; }
+
+}}