New Features:
Multi-Channel and High-Resolution Audio Support
48,000 frames per second ("48k") operation.
48k lossless stereo support.
5.1 and 7.1 surround sound support.
Multi-channel and multi-rate operation on ALSA, PipeWire, PulseAudio, FreeBSD, stdout and Unix pipe output backends.
Automatic Audio Format Selection
Flexible and controllable output format selection.
Automatic rate, sample format, and channel count selection.
Full FFmpeg Integration
Support for transcoding.
Advanced resampling capabilities.
New audio format support.
Enhanced Resampling
New vernier resampling and interpolation method optimized for low-power CPUs.
Better performance on resource-constrained devices.
Convolution and Loudness Enhancements:
Convolution system is now multithreaded and works on stereo and multichannel audio at 48k and 44.1k.
Multiple impulse response (IR) files can now be provided via convolution_ir_files setting.
New convolution_thread_pool_size setting for multithreaded processing (defaults to 1).
Loudness processing now works with stereo and multichannel audio at 48k and 44.1k.
Updated to the most recent HiFi-LoFi FFT convolver.
MQTT Enhancements:
Added new publish_retain boolean option. When enabled, published MQTT messages have the retain flag set, so the MQTT broker stores the last message per topic and new subscribers receive the most recent value immediately. Thanks to lululombard for PR #2142.
D-Bus Enhancements:
Added new dbus_default_message_bus command-line argument (can be system or session) to set the default message bus for both D-Bus native service and MPRIS service.
Performance Improvements:
Enhanced compatibility with AirPlay 2 AutoMix and Smart Tracklists resulting in less unexplained track skipping.
Better operation on low-power devices down to Raspberry Pi B.
Improved efficiency on embedded systems.
Enhanced timestamp handling for better synchronization.
Improved sync error calculation.
Rebuilt buffered audio processor for cleaner handling of immediate and deferred flush requests.
Docker Enhancements:
Reduced Docker image sizes with slimmed-down FFmpeg library.
Removed dhclient from Docker images for smaller footprint.
Bug Fixes:
Fixed MQTT warning on service startup: "Could not establish a mqtt connection". The startup script now correctly states that the mosquitto service is required. Thanks to Hugo Villeneuve for PR #2137.
Fixed compatibility with mbedtls library version 3.4+ (present on recent Linux versions). Thanks to Christian Beier for finding and fixing the bug.
Fixed PulseAudio backend so that PA_ERR_NODATA returns "No latency data yet". Thanks to Vladimir Shakov for the report and fix.
Ensured old flush requests are deleted when a new play session starts. Thanks to saujanyashah for the report.
Fixed format warnings on 64-bit and 32-bit systems
Removed compilation warnings on 32-bit builds
Improved argument checking for debug(), inform(), warn() and die() functions
Fixed "daemon" typos throughout codebase. Thanks to Chris Boot for PR #1981.
Added warning if a convolution impulse response file cannot be read due to bad path or permissions
Build System Improvements:
Unified service file with variable substitution for Avahi support, making it easier to add future service dependencies. Thanks to Hugo Villeneuve.
Network interface selection now only considers interfaces that are up, running and not loopback interfaces. Thanks to Carl Johnson for the suggestion.
Configuration File Changes and Deprecations
New settings: convolution_ir_files (replaces convolution_ir_file), convolution_enabled (replaces convolution), convolution_max_length_in_seconds (replaces convolution_max_length), loudness_enabled (replaces loudness).
New convolution_thread_pool_size setting (defaults to 1).
Deprecated settings: convolution_ir_file, convolution, convolution_max_length, loudness.
Corresponding D-Bus methods and properties have been updated.
Deprecation Notice:
The Jack Audio and soundio backends are deprecated and will be removed in a future release. Consider using the updated PipeWire backend instead.
Documentation Updates
Updated BUILD.md with latest build instructions.
Updated AIRPLAY2.md with feature information.
Enhanced convolution and loudness documentation.
Maintenance:
Fixed FFmpeg deprecation warnings.
Bumped actions/checkout from 6.0.1 to 6.0.2.
Bumped docker/login-action from 3.6.0 to 3.7.0.
Bumped docker/build-push-action from 6.13.0 to 6.15.0.
Bumped docker/setup-qemu-action from 3.4.0 to 3.6.0.
Bumped docker/setup-buildx-action from 3.9.0 to 3.10.0.
223 lines
5.3 KiB
C++
Executable File
223 lines
5.3 KiB
C++
Executable File
// ==================================================================================
|
|
// Copyright (c) 2012 HiFi-LoFi
|
|
//
|
|
// This 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.
|
|
//
|
|
// This program 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 this program. If not, see <http://www.gnu.org/licenses/>.
|
|
// ==================================================================================
|
|
|
|
#include "FFTConvolver.h"
|
|
|
|
#include <cassert>
|
|
#include <cmath>
|
|
|
|
#if defined (FFTCONVOLVER_USE_SSE)
|
|
#include <xmmintrin.h>
|
|
#endif
|
|
|
|
|
|
namespace fftconvolver
|
|
{
|
|
|
|
FFTConvolver::FFTConvolver() :
|
|
_blockSize(0),
|
|
_segSize(0),
|
|
_segCount(0),
|
|
_fftComplexSize(0),
|
|
_segments(),
|
|
_segmentsIR(),
|
|
_fftBuffer(),
|
|
_fft(),
|
|
_preMultiplied(),
|
|
_conv(),
|
|
_overlap(),
|
|
_current(0),
|
|
_inputBuffer(),
|
|
_inputBufferFill(0)
|
|
{
|
|
}
|
|
|
|
|
|
FFTConvolver::~FFTConvolver()
|
|
{
|
|
reset();
|
|
}
|
|
|
|
|
|
void FFTConvolver::reset()
|
|
{
|
|
for (size_t i=0; i<_segCount; ++i)
|
|
{
|
|
delete _segments[i];
|
|
delete _segmentsIR[i];
|
|
}
|
|
|
|
_blockSize = 0;
|
|
_segSize = 0;
|
|
_segCount = 0;
|
|
_fftComplexSize = 0;
|
|
_segments.clear();
|
|
_segmentsIR.clear();
|
|
_fftBuffer.clear();
|
|
_fft.init(0);
|
|
_preMultiplied.clear();
|
|
_conv.clear();
|
|
_overlap.clear();
|
|
_current = 0;
|
|
_inputBuffer.clear();
|
|
_inputBufferFill = 0;
|
|
}
|
|
|
|
void FFTConvolver::clearState()
|
|
{
|
|
if (_segCount == 0)
|
|
{
|
|
return; // Not initialized
|
|
}
|
|
|
|
_inputBuffer.setZero();
|
|
_inputBufferFill = 0;
|
|
_overlap.setZero();
|
|
|
|
for (size_t i = 0; i < _segCount; ++i)
|
|
{
|
|
_segments[i]->setZero();
|
|
}
|
|
|
|
_preMultiplied.setZero();
|
|
_conv.setZero();
|
|
_current = 0;
|
|
}
|
|
|
|
bool FFTConvolver::init(size_t blockSize, const Sample* ir, size_t irLen)
|
|
{
|
|
reset();
|
|
|
|
if (blockSize == 0)
|
|
{
|
|
return false;
|
|
}
|
|
|
|
// Ignore zeros at the end of the impulse response because they only waste computation time
|
|
while (irLen > 0 && ::fabs(ir[irLen-1]) < 0.000001f)
|
|
{
|
|
--irLen;
|
|
}
|
|
|
|
if (irLen == 0)
|
|
{
|
|
return true;
|
|
}
|
|
|
|
_blockSize = NextPowerOf2(blockSize);
|
|
_segSize = 2 * _blockSize;
|
|
_segCount = static_cast<size_t>(::ceil(static_cast<float>(irLen) / static_cast<float>(_blockSize)));
|
|
_fftComplexSize = audiofft::AudioFFT::ComplexSize(_segSize);
|
|
|
|
// FFT
|
|
_fft.init(_segSize);
|
|
_fftBuffer.resize(_segSize);
|
|
|
|
// Prepare segments
|
|
for (size_t i=0; i<_segCount; ++i)
|
|
{
|
|
_segments.push_back(new SplitComplex(_fftComplexSize));
|
|
}
|
|
|
|
// Prepare IR
|
|
for (size_t i=0; i<_segCount; ++i)
|
|
{
|
|
SplitComplex* segment = new SplitComplex(_fftComplexSize);
|
|
const size_t remaining = irLen - (i * _blockSize);
|
|
const size_t sizeCopy = (remaining >= _blockSize) ? _blockSize : remaining;
|
|
CopyAndPad(_fftBuffer, &ir[i*_blockSize], sizeCopy);
|
|
_fft.fft(_fftBuffer.data(), segment->re(), segment->im());
|
|
_segmentsIR.push_back(segment);
|
|
}
|
|
|
|
// Prepare convolution buffers
|
|
_preMultiplied.resize(_fftComplexSize);
|
|
_conv.resize(_fftComplexSize);
|
|
_overlap.resize(_blockSize);
|
|
|
|
// Prepare input buffer
|
|
_inputBuffer.resize(_blockSize);
|
|
_inputBufferFill = 0;
|
|
|
|
// Reset current position
|
|
_current = 0;
|
|
|
|
return true;
|
|
}
|
|
|
|
|
|
void FFTConvolver::process(const Sample* input, Sample* output, size_t len)
|
|
{
|
|
if (_segCount == 0)
|
|
{
|
|
::memset(output, 0, len * sizeof(Sample));
|
|
return;
|
|
}
|
|
|
|
size_t processed = 0;
|
|
while (processed < len)
|
|
{
|
|
const bool inputBufferWasEmpty = (_inputBufferFill == 0);
|
|
const size_t processing = std::min(len-processed, _blockSize-_inputBufferFill);
|
|
const size_t inputBufferPos = _inputBufferFill;
|
|
::memcpy(_inputBuffer.data()+inputBufferPos, input+processed, processing * sizeof(Sample));
|
|
|
|
// Forward FFT
|
|
CopyAndPad(_fftBuffer, &_inputBuffer[0], _blockSize);
|
|
_fft.fft(_fftBuffer.data(), _segments[_current]->re(), _segments[_current]->im());
|
|
|
|
// Complex multiplication
|
|
if (inputBufferWasEmpty)
|
|
{
|
|
_preMultiplied.setZero();
|
|
for (size_t i=1; i<_segCount; ++i)
|
|
{
|
|
const size_t indexIr = i;
|
|
const size_t indexAudio = (_current + i) % _segCount;
|
|
ComplexMultiplyAccumulate(_preMultiplied, *_segmentsIR[indexIr], *_segments[indexAudio]);
|
|
}
|
|
}
|
|
_conv.copyFrom(_preMultiplied);
|
|
ComplexMultiplyAccumulate(_conv, *_segments[_current], *_segmentsIR[0]);
|
|
|
|
// Backward FFT
|
|
_fft.ifft(_fftBuffer.data(), _conv.re(), _conv.im());
|
|
|
|
// Add overlap
|
|
Sum(output+processed, _fftBuffer.data()+inputBufferPos, _overlap.data()+inputBufferPos, processing);
|
|
|
|
// Input buffer full => Next block
|
|
_inputBufferFill += processing;
|
|
if (_inputBufferFill == _blockSize)
|
|
{
|
|
// Input buffer is empty again now
|
|
_inputBuffer.setZero();
|
|
_inputBufferFill = 0;
|
|
|
|
// Save the overlap
|
|
::memcpy(_overlap.data(), _fftBuffer.data()+_blockSize, _blockSize * sizeof(Sample));
|
|
|
|
// Update current segment
|
|
_current = (_current > 0) ? (_current - 1) : (_segCount - 1);
|
|
}
|
|
|
|
processed += processing;
|
|
}
|
|
}
|
|
|
|
} // End of namespace fftconvolver
|