chromium/third_party/webrtc/modules/audio_processing/agc2/rnn_vad/spectral_features.cc

/*
 *  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.
 */

#include "modules/audio_processing/agc2/rnn_vad/spectral_features.h"

#include <algorithm>
#include <cmath>
#include <limits>
#include <numeric>

#include "rtc_base/checks.h"
#include "rtc_base/numerics/safe_compare.h"

namespace webrtc {
namespace rnn_vad {
namespace {

constexpr float kSilenceThreshold = …;

// Computes the new cepstral difference stats and pushes them into the passed
// symmetric matrix buffer.
void UpdateCepstralDifferenceStats(
    rtc::ArrayView<const float, kNumBands> new_cepstral_coeffs,
    const RingBuffer<float, kNumBands, kCepstralCoeffsHistorySize>& ring_buf,
    SymmetricMatrixBuffer<float, kCepstralCoeffsHistorySize>* sym_matrix_buf) { … }

// Computes the first half of the Vorbis window.
std::array<float, kFrameSize20ms24kHz / 2> ComputeScaledHalfVorbisWindow(
    float scaling = 1.f) { … }

// Computes the forward FFT on a 20 ms frame to which a given window function is
// applied. The Fourier coefficient corresponding to the Nyquist frequency is
// set to zero (it is never used and this allows to simplify the code).
void ComputeWindowedForwardFft(
    rtc::ArrayView<const float, kFrameSize20ms24kHz> frame,
    const std::array<float, kFrameSize20ms24kHz / 2>& half_window,
    Pffft::FloatBuffer* fft_input_buffer,
    Pffft::FloatBuffer* fft_output_buffer,
    Pffft* fft) { … }

}  // namespace

SpectralFeaturesExtractor::SpectralFeaturesExtractor()
    : … { … }

SpectralFeaturesExtractor::~SpectralFeaturesExtractor() = default;

void SpectralFeaturesExtractor::Reset() { … }

bool SpectralFeaturesExtractor::CheckSilenceComputeFeatures(
    rtc::ArrayView<const float, kFrameSize20ms24kHz> reference_frame,
    rtc::ArrayView<const float, kFrameSize20ms24kHz> lagged_frame,
    rtc::ArrayView<float, kNumBands - kNumLowerBands> higher_bands_cepstrum,
    rtc::ArrayView<float, kNumLowerBands> average,
    rtc::ArrayView<float, kNumLowerBands> first_derivative,
    rtc::ArrayView<float, kNumLowerBands> second_derivative,
    rtc::ArrayView<float, kNumLowerBands> bands_cross_corr,
    float* variability) { … }

void SpectralFeaturesExtractor::ComputeAvgAndDerivatives(
    rtc::ArrayView<float, kNumLowerBands> average,
    rtc::ArrayView<float, kNumLowerBands> first_derivative,
    rtc::ArrayView<float, kNumLowerBands> second_derivative) const { … }

void SpectralFeaturesExtractor::ComputeNormalizedCepstralCorrelation(
    rtc::ArrayView<float, kNumLowerBands> bands_cross_corr) { … }

float SpectralFeaturesExtractor::ComputeVariability() const { … }

}  // namespace rnn_vad
}  // namespace webrtc