| /* |
| * Copyright (C) 2011 Google Inc. All rights reserved. |
| * |
| * Redistribution and use in source and binary forms, with or without |
| * modification, are permitted provided that the following conditions |
| * are met: |
| * |
| * 1. Redistributions of source code must retain the above copyright |
| * notice, this list of conditions and the following disclaimer. |
| * 2. Redistributions in binary form must reproduce the above copyright |
| * notice, this list of conditions and the following disclaimer in the |
| * documentation and/or other materials provided with the distribution. |
| * 3. Neither the name of Apple Inc. ("Apple") nor the names of |
| * its contributors may be used to endorse or promote products derived |
| * from this software without specific prior written permission. |
| * |
| * THIS SOFTWARE IS PROVIDED BY APPLE AND ITS CONTRIBUTORS "AS IS" AND ANY |
| * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED |
| * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE |
| * DISCLAIMED. IN NO EVENT SHALL APPLE OR ITS CONTRIBUTORS BE LIABLE FOR ANY |
| * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES |
| * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; |
| * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND |
| * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
| * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF |
| * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
| */ |
| |
| #include "config.h" |
| |
| #if ENABLE(WEB_AUDIO) |
| |
| #include "DynamicsCompressorKernel.h" |
| |
| #include "AudioUtilities.h" |
| #include "DenormalDisabler.h" |
| #include <algorithm> |
| #include <wtf/MathExtras.h> |
| |
| namespace WebCore { |
| |
| using namespace AudioUtilities; |
| |
| // Metering hits peaks instantly, but releases this fast (in seconds). |
| constexpr float meteringReleaseTimeConstant = 0.325f; |
| |
| DynamicsCompressorKernel::DynamicsCompressorKernel(float sampleRate, unsigned numberOfChannels) |
| : m_sampleRate(sampleRate) |
| { |
| setNumberOfChannels(numberOfChannels); |
| |
| // Initializes most member variables |
| reset(); |
| |
| m_meteringReleaseK = static_cast<float>(discreteTimeConstantForSampleRate(meteringReleaseTimeConstant, sampleRate)); |
| } |
| |
| void DynamicsCompressorKernel::setNumberOfChannels(unsigned numberOfChannels) |
| { |
| if (m_preDelayBuffers.size() == numberOfChannels) |
| return; |
| |
| m_preDelayBuffers.clear(); |
| m_preDelayBuffers.reserveInitialCapacity(numberOfChannels); |
| for (unsigned i = 0; i < numberOfChannels; ++i) |
| m_preDelayBuffers.uncheckedAppend(makeUnique<AudioFloatArray>(MaxPreDelayFrames)); |
| } |
| |
| void DynamicsCompressorKernel::setPreDelayTime(float preDelayTime) |
| { |
| // Re-configure look-ahead section pre-delay if delay time has changed. |
| unsigned preDelayFrames = preDelayTime * sampleRate(); |
| if (preDelayFrames > MaxPreDelayFrames - 1) |
| preDelayFrames = MaxPreDelayFrames - 1; |
| |
| if (m_lastPreDelayFrames != preDelayFrames) { |
| m_lastPreDelayFrames = preDelayFrames; |
| for (unsigned i = 0; i < m_preDelayBuffers.size(); ++i) |
| m_preDelayBuffers[i]->zero(); |
| |
| m_preDelayReadIndex = 0; |
| m_preDelayWriteIndex = preDelayFrames; |
| } |
| } |
| |
| // Exponential curve for the knee. |
| // It is 1st derivative matched at m_linearThreshold and asymptotically approaches the value m_linearThreshold + 1 / k. |
| float DynamicsCompressorKernel::kneeCurve(float x, float k) |
| { |
| // Linear up to threshold. |
| if (x < m_linearThreshold) |
| return x; |
| |
| return m_linearThreshold + (1 - expf(-k * (x - m_linearThreshold))) / k; |
| } |
| |
| // Full compression curve with constant ratio after knee. |
| float DynamicsCompressorKernel::saturate(float x, float k) |
| { |
| float y; |
| |
| if (x < m_kneeThreshold) |
| y = kneeCurve(x, k); |
| else { |
| // Constant ratio after knee. |
| float xDb = linearToDecibels(x); |
| float yDb = m_ykneeThresholdDb + m_slope * (xDb - m_kneeThresholdDb); |
| |
| y = decibelsToLinear(yDb); |
| } |
| |
| return y; |
| } |
| |
| // Approximate 1st derivative with input and output expressed in dB. |
| // This slope is equal to the inverse of the compression "ratio". |
| // In other words, a compression ratio of 20 would be a slope of 1/20. |
| float DynamicsCompressorKernel::slopeAt(float x, float k) |
| { |
| if (x < m_linearThreshold) |
| return 1; |
| |
| float x2 = x * 1.001; |
| |
| float xDb = linearToDecibels(x); |
| float x2Db = linearToDecibels(x2); |
| |
| float yDb = linearToDecibels(kneeCurve(x, k)); |
| float y2Db = linearToDecibels(kneeCurve(x2, k)); |
| |
| float m = (y2Db - yDb) / (x2Db - xDb); |
| |
| return m; |
| } |
| |
| float DynamicsCompressorKernel::kAtSlope(float desiredSlope) |
| { |
| float xDb = m_dbThreshold + m_dbKnee; |
| float x = decibelsToLinear(xDb); |
| |
| // Approximate k given initial values. |
| float minK = 0.1; |
| float maxK = 10000; |
| float k = 5; |
| |
| for (int i = 0; i < 15; ++i) { |
| // A high value for k will more quickly asymptotically approach a slope of 0. |
| float slope = slopeAt(x, k); |
| |
| if (slope < desiredSlope) { |
| // k is too high. |
| maxK = k; |
| } else { |
| // k is too low. |
| minK = k; |
| } |
| |
| // Re-calculate based on geometric mean. |
| k = sqrtf(minK * maxK); |
| } |
| |
| return k; |
| } |
| |
| float DynamicsCompressorKernel::updateStaticCurveParameters(float dbThreshold, float dbKnee, float ratio) |
| { |
| if (dbThreshold != m_dbThreshold || dbKnee != m_dbKnee || ratio != m_ratio) { |
| // Threshold and knee. |
| m_dbThreshold = dbThreshold; |
| m_linearThreshold = decibelsToLinear(dbThreshold); |
| m_dbKnee = dbKnee; |
| |
| // Compute knee parameters. |
| m_ratio = ratio; |
| m_slope = 1 / m_ratio; |
| |
| float k = kAtSlope(1 / m_ratio); |
| |
| m_kneeThresholdDb = dbThreshold + dbKnee; |
| m_kneeThreshold = decibelsToLinear(m_kneeThresholdDb); |
| |
| m_ykneeThresholdDb = linearToDecibels(kneeCurve(m_kneeThreshold, k)); |
| |
| m_K = k; |
| } |
| return m_K; |
| } |
| |
| void DynamicsCompressorKernel::process(const float* sourceChannels[], |
| float* destinationChannels[], |
| unsigned numberOfChannels, |
| unsigned framesToProcess, |
| |
| float dbThreshold, |
| float dbKnee, |
| float ratio, |
| float attackTime, |
| float releaseTime, |
| float preDelayTime, |
| float dbPostGain, |
| float effectBlend, /* equal power crossfade */ |
| |
| float releaseZone1, |
| float releaseZone2, |
| float releaseZone3, |
| float releaseZone4 |
| ) |
| { |
| ASSERT(m_preDelayBuffers.size() == numberOfChannels); |
| |
| float sampleRate = this->sampleRate(); |
| |
| float dryMix = 1 - effectBlend; |
| float wetMix = effectBlend; |
| |
| float k = updateStaticCurveParameters(dbThreshold, dbKnee, ratio); |
| |
| // Makeup gain. |
| float fullRangeGain = saturate(1, k); |
| float fullRangeMakeupGain = 1 / fullRangeGain; |
| |
| // Empirical/perceptual tuning. |
| fullRangeMakeupGain = powf(fullRangeMakeupGain, 0.6f); |
| |
| float masterLinearGain = decibelsToLinear(dbPostGain) * fullRangeMakeupGain; |
| |
| // Attack parameters. |
| attackTime = std::max(0.001f, attackTime); |
| float attackFrames = attackTime * sampleRate; |
| |
| // Release parameters. |
| float releaseFrames = sampleRate * releaseTime; |
| |
| // Detector release time. |
| float satReleaseTime = 0.0025f; |
| float satReleaseFrames = satReleaseTime * sampleRate; |
| |
| // Create a smooth function which passes through four points. |
| |
| // Polynomial of the form |
| // y = a + b*x + c*x^2 + d*x^3 + e*x^4; |
| |
| float y1 = releaseFrames * releaseZone1; |
| float y2 = releaseFrames * releaseZone2; |
| float y3 = releaseFrames * releaseZone3; |
| float y4 = releaseFrames * releaseZone4; |
| |
| // All of these coefficients were derived for 4th order polynomial curve fitting where the y values |
| // match the evenly spaced x values as follows: (y1 : x == 0, y2 : x == 1, y3 : x == 2, y4 : x == 3) |
| float kA = 0.9999999999999998f*y1 + 1.8432219684323923e-16f*y2 - 1.9373394351676423e-16f*y3 + 8.824516011816245e-18f*y4; |
| float kB = -1.5788320352845888f*y1 + 2.3305837032074286f*y2 - 0.9141194204840429f*y3 + 0.1623677525612032f*y4; |
| float kC = 0.5334142869106424f*y1 - 1.272736789213631f*y2 + 0.9258856042207512f*y3 - 0.18656310191776226f*y4; |
| float kD = 0.08783463138207234f*y1 - 0.1694162967925622f*y2 + 0.08588057951595272f*y3 - 0.00429891410546283f*y4; |
| float kE = -0.042416883008123074f*y1 + 0.1115693827987602f*y2 - 0.09764676325265872f*y3 + 0.028494263462021576f*y4; |
| |
| // x ranges from 0 -> 3 0 1 2 3 |
| // -15 -10 -5 0db |
| |
| // y calculates adaptive release frames depending on the amount of compression. |
| |
| setPreDelayTime(preDelayTime); |
| |
| const int nDivisionFrames = 32; |
| |
| const int nDivisions = framesToProcess / nDivisionFrames; |
| |
| unsigned frameIndex = 0; |
| for (int i = 0; i < nDivisions; ++i) { |
| // ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
| // Calculate desired gain |
| // ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
| |
| // Fix gremlins. |
| if (std::isnan(m_detectorAverage)) |
| m_detectorAverage = 1; |
| if (std::isinf(m_detectorAverage)) |
| m_detectorAverage = 1; |
| |
| float desiredGain = m_detectorAverage; |
| |
| // Pre-warp so we get desiredGain after sin() warp below. |
| float scaledDesiredGain = asinf(desiredGain) / (0.5f * piFloat); |
| |
| // ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
| // Deal with envelopes |
| // ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
| |
| // envelopeRate is the rate we slew from current compressor level to the desired level. |
| // The exact rate depends on if we're attacking or releasing and by how much. |
| float envelopeRate; |
| |
| bool isReleasing = scaledDesiredGain > m_compressorGain; |
| |
| // compressionDiffDb is the difference between current compression level and the desired level. |
| float compressionDiffDb = linearToDecibels(m_compressorGain / scaledDesiredGain); |
| |
| if (isReleasing) { |
| // Release mode - compressionDiffDb should be negative dB |
| m_maxAttackCompressionDiffDb = -1; |
| |
| // Fix gremlins. |
| if (std::isnan(compressionDiffDb)) |
| compressionDiffDb = -1; |
| if (std::isinf(compressionDiffDb)) |
| compressionDiffDb = -1; |
| |
| // Adaptive release - higher compression (lower compressionDiffDb) releases faster. |
| |
| // Contain within range: -12 -> 0 then scale to go from 0 -> 3 |
| float x = compressionDiffDb; |
| x = std::max(-12.0f, x); |
| x = std::min(0.0f, x); |
| x = 0.25f * (x + 12); |
| |
| // Compute adaptive release curve using 4th order polynomial. |
| // Normal values for the polynomial coefficients would create a monotonically increasing function. |
| float x2 = x * x; |
| float x3 = x2 * x; |
| float x4 = x2 * x2; |
| float releaseFrames = kA + kB * x + kC * x2 + kD * x3 + kE * x4; |
| |
| #define kSpacingDb 5 |
| float dbPerFrame = kSpacingDb / releaseFrames; |
| |
| envelopeRate = decibelsToLinear(dbPerFrame); |
| } else { |
| // Attack mode - compressionDiffDb should be positive dB |
| |
| // Fix gremlins. |
| if (std::isnan(compressionDiffDb)) |
| compressionDiffDb = 1; |
| if (std::isinf(compressionDiffDb)) |
| compressionDiffDb = 1; |
| |
| // As long as we're still in attack mode, use a rate based off |
| // the largest compressionDiffDb we've encountered so far. |
| if (m_maxAttackCompressionDiffDb == -1 || m_maxAttackCompressionDiffDb < compressionDiffDb) |
| m_maxAttackCompressionDiffDb = compressionDiffDb; |
| |
| float effAttenDiffDb = std::max(0.5f, m_maxAttackCompressionDiffDb); |
| |
| float x = 0.25f / effAttenDiffDb; |
| envelopeRate = 1 - powf(x, 1 / attackFrames); |
| } |
| |
| // ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
| // Inner loop - calculate shaped power average - apply compression. |
| // ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
| |
| { |
| int preDelayReadIndex = m_preDelayReadIndex; |
| int preDelayWriteIndex = m_preDelayWriteIndex; |
| float detectorAverage = m_detectorAverage; |
| float compressorGain = m_compressorGain; |
| |
| int loopFrames = nDivisionFrames; |
| while (loopFrames--) { |
| float compressorInput = 0; |
| |
| // Predelay signal, computing compression amount from un-delayed version. |
| for (unsigned i = 0; i < numberOfChannels; ++i) { |
| float* delayBuffer = m_preDelayBuffers[i]->data(); |
| float undelayedSource = sourceChannels[i][frameIndex]; |
| delayBuffer[preDelayWriteIndex] = undelayedSource; |
| |
| float absUndelayedSource = undelayedSource > 0 ? undelayedSource : -undelayedSource; |
| if (compressorInput < absUndelayedSource) |
| compressorInput = absUndelayedSource; |
| } |
| |
| // Calculate shaped power on undelayed input. |
| |
| float scaledInput = compressorInput; |
| float absInput = scaledInput > 0 ? scaledInput : -scaledInput; |
| |
| // Put through shaping curve. |
| // This is linear up to the threshold, then enters a "knee" portion followed by the "ratio" portion. |
| // The transition from the threshold to the knee is smooth (1st derivative matched). |
| // The transition from the knee to the ratio portion is smooth (1st derivative matched). |
| float shapedInput = saturate(absInput, k); |
| |
| float attenuation = absInput <= 0.0001f ? 1 : shapedInput / absInput; |
| |
| float attenuationDb = -linearToDecibels(attenuation); |
| attenuationDb = std::max(2.0f, attenuationDb); |
| |
| float dbPerFrame = attenuationDb / satReleaseFrames; |
| |
| float satReleaseRate = decibelsToLinear(dbPerFrame) - 1; |
| |
| bool isRelease = (attenuation > detectorAverage); |
| float rate = isRelease ? satReleaseRate : 1; |
| |
| detectorAverage += (attenuation - detectorAverage) * rate; |
| detectorAverage = std::min(1.0f, detectorAverage); |
| |
| // Fix gremlins. |
| if (std::isnan(detectorAverage)) |
| detectorAverage = 1; |
| if (std::isinf(detectorAverage)) |
| detectorAverage = 1; |
| |
| // Exponential approach to desired gain. |
| if (envelopeRate < 1) { |
| // Attack - reduce gain to desired. |
| compressorGain += (scaledDesiredGain - compressorGain) * envelopeRate; |
| } else { |
| // Release - exponentially increase gain to 1.0 |
| compressorGain *= envelopeRate; |
| compressorGain = std::min(1.0f, compressorGain); |
| } |
| |
| // Warp pre-compression gain to smooth out sharp exponential transition points. |
| float postWarpCompressorGain = sinf(0.5f * piFloat * compressorGain); |
| |
| // Calculate total gain using master gain and effect blend. |
| float totalGain = dryMix + wetMix * masterLinearGain * postWarpCompressorGain; |
| |
| // Calculate metering. |
| float dbRealGain = 20 * log10(postWarpCompressorGain); |
| if (dbRealGain < m_meteringGain) |
| m_meteringGain = dbRealGain; |
| else |
| m_meteringGain += (dbRealGain - m_meteringGain) * m_meteringReleaseK; |
| |
| // Apply final gain. |
| for (unsigned i = 0; i < numberOfChannels; ++i) { |
| float* delayBuffer = m_preDelayBuffers[i]->data(); |
| destinationChannels[i][frameIndex] = delayBuffer[preDelayReadIndex] * totalGain; |
| } |
| |
| frameIndex++; |
| preDelayReadIndex = (preDelayReadIndex + 1) & MaxPreDelayFramesMask; |
| preDelayWriteIndex = (preDelayWriteIndex + 1) & MaxPreDelayFramesMask; |
| } |
| |
| // Locals back to member variables. |
| m_preDelayReadIndex = preDelayReadIndex; |
| m_preDelayWriteIndex = preDelayWriteIndex; |
| m_detectorAverage = DenormalDisabler::flushDenormalFloatToZero(detectorAverage); |
| m_compressorGain = DenormalDisabler::flushDenormalFloatToZero(compressorGain); |
| } |
| } |
| } |
| |
| void DynamicsCompressorKernel::reset() |
| { |
| m_detectorAverage = 0; |
| m_compressorGain = 1; |
| m_meteringGain = 1; |
| |
| // Predelay section. |
| for (unsigned i = 0; i < m_preDelayBuffers.size(); ++i) |
| m_preDelayBuffers[i]->zero(); |
| |
| m_preDelayReadIndex = 0; |
| m_preDelayWriteIndex = DefaultPreDelayFrames; |
| |
| m_maxAttackCompressionDiffDb = -1; // uninitialized state |
| } |
| |
| double DynamicsCompressorKernel::tailTime() const |
| { |
| // The reduction value of the compressor is computed from the gain using an exponential filter |
| // with a time constant of |meteringReleaseTimeConstant|. We need to keep he compressor running |
| // for some time after the inputs go away so that the reduction value approaches 0. This is a |
| // tradeoff between how long we keep the node alive and how close we approach the final value. |
| // A value of 5 to 10 times the time constant is a reasonable trade-off. |
| return 5 * meteringReleaseTimeConstant; |
| } |
| |
| } // namespace WebCore |
| |
| #endif // ENABLE(WEB_AUDIO) |