980c4601e1
- Bug fix: the desired initial gain quickly dropped to 0 dB hence starting a call with a too low level - New tuning to make AGC2 more robust to VAD mistakes - Smarter max gain increase speed: to deal with an increased threshold of adjacent speech frames, the gain applier temporarily allows a faster gain increase to deal with a longer time spent waiting for enough speech frames in a row to be observed - Saturation protector isolated from `AdaptiveModeLevelEstimator` to simplify the unit tests for the latter (non bit-exact change) - AGC2 adaptive digital config: unnecessary params deprecated - Code readability improvements - Data dumps clean-up and better naming Bug: webrtc:7494 Change-Id: I4e36059bdf2566cc2a7e1a7e95b7430ba9ae9844 Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/215140 Commit-Queue: Alessio Bazzica <alessiob@webrtc.org> Reviewed-by: Jesus de Vicente Pena <devicentepena@webrtc.org> Cr-Commit-Position: refs/heads/master@{#33736}
69 lines
2.8 KiB
C++
69 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_ADAPTIVE_DIGITAL_GAIN_APPLIER_H_
|
|
#define MODULES_AUDIO_PROCESSING_AGC2_ADAPTIVE_DIGITAL_GAIN_APPLIER_H_
|
|
|
|
#include "modules/audio_processing/agc2/gain_applier.h"
|
|
#include "modules/audio_processing/include/audio_frame_view.h"
|
|
|
|
namespace webrtc {
|
|
|
|
class ApmDataDumper;
|
|
|
|
// TODO(bugs.webrtc.org): Split into `GainAdaptor` and `GainApplier`.
|
|
// Selects the target digital gain, decides when and how quickly to adapt to the
|
|
// target and applies the current gain to 10 ms frames.
|
|
class AdaptiveDigitalGainApplier {
|
|
public:
|
|
// Information about a frame to process.
|
|
struct FrameInfo {
|
|
float speech_probability; // Probability of speech in the [0, 1] range.
|
|
float speech_level_dbfs; // Estimated speech level (dBFS).
|
|
bool speech_level_reliable; // True with reliable speech level estimation.
|
|
float noise_rms_dbfs; // Estimated noise RMS level (dBFS).
|
|
float headroom_db; // Headroom (dB).
|
|
float limiter_envelope_dbfs; // Envelope level from the limiter (dBFS).
|
|
};
|
|
|
|
// Ctor. `adjacent_speech_frames_threshold` indicates how many adjacent speech
|
|
// frames must be observed in order to consider the sequence as speech.
|
|
// `max_gain_change_db_per_second` limits the adaptation speed (uniformly
|
|
// operated across frames). `max_output_noise_level_dbfs` limits the output
|
|
// noise level.
|
|
AdaptiveDigitalGainApplier(ApmDataDumper* apm_data_dumper,
|
|
int adjacent_speech_frames_threshold,
|
|
float max_gain_change_db_per_second,
|
|
float max_output_noise_level_dbfs);
|
|
AdaptiveDigitalGainApplier(const AdaptiveDigitalGainApplier&) = delete;
|
|
AdaptiveDigitalGainApplier& operator=(const AdaptiveDigitalGainApplier&) =
|
|
delete;
|
|
|
|
// Analyzes `info`, updates the digital gain and applies it to a 10 ms
|
|
// `frame`. Supports any sample rate supported by APM.
|
|
void Process(const FrameInfo& info, AudioFrameView<float> frame);
|
|
|
|
private:
|
|
ApmDataDumper* const apm_data_dumper_;
|
|
GainApplier gain_applier_;
|
|
|
|
const int adjacent_speech_frames_threshold_;
|
|
const float max_gain_change_db_per_10ms_;
|
|
const float max_output_noise_level_dbfs_;
|
|
|
|
int calls_since_last_gain_log_;
|
|
int frames_to_gain_increase_allowed_;
|
|
float last_gain_db_;
|
|
};
|
|
|
|
} // namespace webrtc
|
|
|
|
#endif // MODULES_AUDIO_PROCESSING_AGC2_ADAPTIVE_DIGITAL_GAIN_APPLIER_H_
|