Alessio Bazzica 5da581b564 AGC2: use only one headroom parameter
Instead of using two different headroom parameters, namely
`kHeadroomDbfs` and `kSaturationProtectorExtraHeadroomDb`, only use
the former that now also accounts for the deleted one - i.e., it equals
the sum of the two headrooms. In this way, tuning AGC2 will be easier.

This CL does *not* change the behavior of the AGC2 adaptive digital
controller - bitexactness verified with audioproc_f on a collection of
AEC dumps and Wav files (42 recordings in total).

The unit tests changes in agc2/saturation_protector_unittest.cc are
required since `extra_headroom_db` is removed and the changes in
agc2/adaptive_digital_gain_applier_unittest.cc are required because
`AdaptiveDigitalGainApplier` depends on `kHeadroomDbfs` which has been
updated as stated above.

Bug: webrtc:7494
Change-Id: I0a2a710bbede0caa53938090a004d185fdefaeb9
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/232905
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#35109}
2021-09-28 16:52:16 +00:00

70 lines
2.8 KiB
C++

/*
* Copyright (c) 2018 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.
*/
#ifndef MODULES_AUDIO_PROCESSING_AGC2_AGC2_COMMON_H_
#define MODULES_AUDIO_PROCESSING_AGC2_AGC2_COMMON_H_
namespace webrtc {
constexpr float kMinFloatS16Value = -32768.0f;
constexpr float kMaxFloatS16Value = 32767.0f;
constexpr float kMaxAbsFloatS16Value = 32768.0f;
// Minimum audio level in dBFS scale for S16 samples.
constexpr float kMinLevelDbfs = -90.31f;
constexpr int kFrameDurationMs = 10;
constexpr int kSubFramesInFrame = 20;
constexpr int kMaximalNumberOfSamplesPerChannel = 480;
// Adaptive digital gain applier settings below.
constexpr float kHeadroomDbfs = 6.0f;
constexpr float kMaxGainDb = 30.0f;
constexpr float kInitialAdaptiveDigitalGainDb = 8.0f;
// At what limiter levels should we start decreasing the adaptive digital gain.
constexpr float kLimiterThresholdForAgcGainDbfs = -1.0f;
// This is the threshold for speech. Speech frames are used for updating the
// speech level, measuring the amount of speech, and decide when to allow target
// gain reduction.
constexpr float kVadConfidenceThreshold = 0.95f;
// Adaptive digital level estimator parameters.
// Number of milliseconds of speech frames to observe to make the estimator
// confident.
constexpr float kLevelEstimatorTimeToConfidenceMs = 400;
constexpr float kLevelEstimatorLeakFactor =
1.0f - 1.0f / kLevelEstimatorTimeToConfidenceMs;
// Robust VAD probability and speech decisions.
constexpr int kDefaultLevelEstimatorAdjacentSpeechFramesThreshold = 12;
// Saturation Protector settings.
constexpr float kSaturationProtectorInitialHeadroomDb = 20.0f;
constexpr int kSaturationProtectorBufferSize = 4;
// Set the initial speech level estimate so that `kInitialAdaptiveDigitalGainDb`
// is applied at the beginning of the call.
constexpr float kInitialSpeechLevelEstimateDbfs =
-kSaturationProtectorInitialHeadroomDb - kInitialAdaptiveDigitalGainDb -
kHeadroomDbfs;
// Number of interpolation points for each region of the limiter.
// These values have been tuned to limit the interpolated gain curve error given
// the limiter parameters and allowing a maximum error of +/- 32768^-1.
constexpr int kInterpolatedGainCurveKneePoints = 22;
constexpr int kInterpolatedGainCurveBeyondKneePoints = 10;
constexpr int kInterpolatedGainCurveTotalPoints =
kInterpolatedGainCurveKneePoints + kInterpolatedGainCurveBeyondKneePoints;
} // namespace webrtc
#endif // MODULES_AUDIO_PROCESSING_AGC2_AGC2_COMMON_H_