d94659dc27
This is the first public upload of the new NetEq, version 4. It has been through extensive internal review during the course of the project. TEST=trybots Review URL: https://webrtc-codereview.appspot.com/1073005 git-svn-id: http://webrtc.googlecode.com/svn/trunk@3425 4adac7df-926f-26a2-2b94-8c16560cd09d
190 lines
8.0 KiB
C++
190 lines
8.0 KiB
C++
/*
|
|
* Copyright (c) 2012 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 "webrtc/modules/audio_coding/neteq4/normal.h"
|
|
|
|
#include <algorithm> // min
|
|
#include <cstring> // memset, memcpy
|
|
|
|
#include "webrtc/common_audio/signal_processing/include/signal_processing_library.h"
|
|
#include "webrtc/modules/audio_coding/codecs/cng/include/webrtc_cng.h"
|
|
#include "webrtc/modules/audio_coding/neteq4/audio_multi_vector.h"
|
|
#include "webrtc/modules/audio_coding/neteq4/background_noise.h"
|
|
#include "webrtc/modules/audio_coding/neteq4/decoder_database.h"
|
|
#include "webrtc/modules/audio_coding/neteq4/expand.h"
|
|
#include "webrtc/modules/audio_coding/neteq4/interface/audio_decoder.h"
|
|
|
|
namespace webrtc {
|
|
|
|
int Normal::Process(const int16_t* input,
|
|
size_t length,
|
|
Modes last_mode,
|
|
int16_t* external_mute_factor_array,
|
|
AudioMultiVector<int16_t>* output) {
|
|
if (length == 0) {
|
|
// Nothing to process.
|
|
output->Clear();
|
|
return length;
|
|
}
|
|
|
|
assert(output->Empty());
|
|
// Output should be empty at this point.
|
|
output->PushBackInterleaved(input, length);
|
|
int16_t* signal = &(*output)[0][0];
|
|
|
|
const unsigned fs_mult = fs_hz_ / 8000;
|
|
assert(fs_mult > 0);
|
|
// fs_shift = log2(fs_mult), rounded down.
|
|
// Note that |fs_shift| is not "exact" for 48 kHz.
|
|
// TODO(hlundin): Investigate this further.
|
|
const int fs_shift = 30 - WebRtcSpl_NormW32(fs_mult);
|
|
|
|
// Check if last RecOut call resulted in an Expand. If so, we have to take
|
|
// care of some cross-fading and unmuting.
|
|
if (last_mode == kModeExpand) {
|
|
// Generate interpolation data using Expand.
|
|
// First, set Expand parameters to appropriate values.
|
|
expand_->SetParametersForNormalAfterExpand();
|
|
|
|
// Call Expand.
|
|
AudioMultiVector<int16_t> expanded(output->Channels());
|
|
expand_->Process(&expanded);
|
|
expand_->Reset();
|
|
|
|
for (size_t channel_ix = 0; channel_ix < output->Channels(); ++channel_ix) {
|
|
// Adjust muting factor (main muting factor times expand muting factor).
|
|
external_mute_factor_array[channel_ix] = static_cast<int16_t>(
|
|
WEBRTC_SPL_MUL_16_16_RSFT(external_mute_factor_array[channel_ix],
|
|
expand_->MuteFactor(channel_ix), 14));
|
|
|
|
int16_t* signal = &(*output)[channel_ix][0];
|
|
size_t length_per_channel = length / output->Channels();
|
|
// Find largest absolute value in new data.
|
|
int16_t decoded_max = WebRtcSpl_MaxAbsValueW16(signal,
|
|
length_per_channel);
|
|
// Adjust muting factor if needed (to BGN level).
|
|
int energy_length = std::min(static_cast<size_t>(fs_mult * 64),
|
|
length_per_channel);
|
|
int scaling = 6 + fs_shift
|
|
- WebRtcSpl_NormW32(decoded_max * decoded_max);
|
|
scaling = std::max(scaling, 0); // |scaling| should always be >= 0.
|
|
int32_t energy = WebRtcSpl_DotProductWithScale(signal, signal,
|
|
energy_length, scaling);
|
|
energy = energy / (energy_length >> scaling);
|
|
|
|
int mute_factor;
|
|
if ((energy != 0) &&
|
|
(energy > background_noise_.Energy(channel_ix))) {
|
|
// Normalize new frame energy to 15 bits.
|
|
scaling = WebRtcSpl_NormW32(energy) - 16;
|
|
// We want background_noise_.energy() / energy in Q14.
|
|
int32_t bgn_energy =
|
|
background_noise_.Energy(channel_ix) << (scaling+14);
|
|
int16_t energy_scaled = energy << scaling;
|
|
int16_t ratio = WebRtcSpl_DivW32W16(bgn_energy, energy_scaled);
|
|
mute_factor = WebRtcSpl_SqrtFloor(static_cast<int32_t>(ratio) << 14);
|
|
} else {
|
|
mute_factor = 16384; // 1.0 in Q14.
|
|
}
|
|
if (mute_factor > external_mute_factor_array[channel_ix]) {
|
|
external_mute_factor_array[channel_ix] = std::min(mute_factor, 16384);
|
|
}
|
|
|
|
// If muted increase by 0.64 for every 20 ms (NB/WB 0.0040/0.0020 in Q14).
|
|
int16_t increment = 64 / fs_mult;
|
|
for (size_t i = 0; i < length_per_channel; i++) {
|
|
// Scale with mute factor.
|
|
assert(channel_ix < output->Channels());
|
|
assert(i < output->Size());
|
|
int32_t scaled_signal = (*output)[channel_ix][i] *
|
|
external_mute_factor_array[channel_ix];
|
|
// Shift 14 with proper rounding.
|
|
(*output)[channel_ix][i] = (scaled_signal + 8192) >> 14;
|
|
// Increase mute_factor towards 16384.
|
|
external_mute_factor_array[channel_ix] =
|
|
std::min(external_mute_factor_array[channel_ix] + increment, 16384);
|
|
}
|
|
|
|
// Interpolate the expanded data into the new vector.
|
|
// (NB/WB/SWB32/SWB48 8/16/32/48 samples.)
|
|
assert(fs_shift < 3); // Will always be 0, 1, or, 2.
|
|
increment = 4 >> fs_shift;
|
|
int fraction = increment;
|
|
for (size_t i = 0; i < 8 * fs_mult; i++) {
|
|
// TODO(hlundin): Add 16 instead of 8 for correct rounding. Keeping 8
|
|
// now for legacy bit-exactness.
|
|
assert(channel_ix < output->Channels());
|
|
assert(i < output->Size());
|
|
(*output)[channel_ix][i] =
|
|
(fraction * (*output)[channel_ix][i] +
|
|
(32 - fraction) * expanded[channel_ix][i] + 8) >> 5;
|
|
fraction += increment;
|
|
}
|
|
}
|
|
} else if (last_mode == kModeRfc3389Cng) {
|
|
assert(output->Channels() == 1); // Not adapted for multi-channel yet.
|
|
static const int kCngLength = 32;
|
|
int16_t cng_output[kCngLength];
|
|
// Reset mute factor and start up fresh.
|
|
external_mute_factor_array[0] = 16384;
|
|
AudioDecoder* cng_decoder = decoder_database_->GetActiveCngDecoder();
|
|
|
|
if (cng_decoder) {
|
|
CNG_dec_inst* cng_inst = static_cast<CNG_dec_inst*>(cng_decoder->state());
|
|
// Generate long enough for 32kHz.
|
|
if (WebRtcCng_Generate(cng_inst, cng_output, kCngLength, 0) < 0) {
|
|
// Error returned; set return vector to all zeros.
|
|
memset(cng_output, 0, sizeof(cng_output));
|
|
}
|
|
} else {
|
|
// If no CNG instance is defined, just copy from the decoded data.
|
|
// (This will result in interpolating the decoded with itself.)
|
|
memcpy(cng_output, signal, fs_mult * 8 * sizeof(int16_t));
|
|
}
|
|
// Interpolate the CNG into the new vector.
|
|
// (NB/WB/SWB32/SWB48 8/16/32/48 samples.)
|
|
assert(fs_shift < 3); // Will always be 0, 1, or, 2.
|
|
int16_t increment = 4 >> fs_shift;
|
|
int16_t fraction = increment;
|
|
for (size_t i = 0; i < 8 * fs_mult; i++) {
|
|
// TODO(hlundin): Add 16 instead of 8 for correct rounding. Keeping 8 now
|
|
// for legacy bit-exactness.
|
|
signal[i] =
|
|
(fraction * signal[i] + (32 - fraction) * cng_output[i] + 8) >> 5;
|
|
fraction += increment;
|
|
}
|
|
} else if (external_mute_factor_array[0] < 16384) {
|
|
// Previous was neither of Expand, FadeToBGN or RFC3389_CNG, but we are
|
|
// still ramping up from previous muting.
|
|
// If muted increase by 0.64 for every 20 ms (NB/WB 0.0040/0.0020 in Q14).
|
|
int16_t increment = 64 / fs_mult;
|
|
size_t length_per_channel = length / output->Channels();
|
|
for (size_t i = 0; i < length_per_channel; i++) {
|
|
for (size_t channel_ix = 0; channel_ix < output->Channels();
|
|
++channel_ix) {
|
|
// Scale with mute factor.
|
|
assert(channel_ix < output->Channels());
|
|
assert(i < output->Size());
|
|
int32_t scaled_signal = (*output)[channel_ix][i] *
|
|
external_mute_factor_array[channel_ix];
|
|
// Shift 14 with proper rounding.
|
|
(*output)[channel_ix][i] = (scaled_signal + 8192) >> 14;
|
|
// Increase mute_factor towards 16384.
|
|
external_mute_factor_array[channel_ix] =
|
|
std::min(16384, external_mute_factor_array[channel_ix] + increment);
|
|
}
|
|
}
|
|
}
|
|
|
|
return length;
|
|
}
|
|
|
|
} // namespace webrtc
|