zqz/platform-external-webrtc
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Reland "RNN VAD: pitch search optimizations (part 3)"

Browse Source 
This reverts commit 57e68ee1b94cce853ab4305680bbe71c01f95e82.

Reason for revert: bug in ancestor CL fixed

Original change's description:
> Revert "RNN VAD: pitch search optimizations (part 3)"
>
> This reverts commit ea89f2a447c514b73da2ed6189fe4b8485f123c6.
>
> Reason for revert: bug in ancestor CL https://webrtc-review.googlesource.com/c/src/+/191320
>
> Original change's description:
> > RNN VAD: pitch search optimizations (part 3)
> >
> > `ComputeSlidingFrameSquareEnergies()` which computes the energy of a
> > sliding 20 ms frame in the pitch buffer has been switched from backward
> > to forward.
> >
> > The benchmark has shown a slight improvement (about +6x).
> >
> > This change is not bit exact but all the tolerance tests still pass
> > except for one single case in `RnnVadTest,PitchSearchWithinTolerance`
> > for which the tolerance has been slightly increased. Note that the pitch
> > estimation is still bit-exact.
> >
> > Benchmarked as follows:
> > ```
> > out/release/modules_unittests \
> >   --gtest_filter=*RnnVadTest.DISABLED_RnnVadPerformance* \
> >   --gtest_also_run_disabled_tests --logs
> > ```
> >
> > Results:
> >
> >       | baseline             | this CL
> > ------+----------------------+------------------------
> > run 1 | 22.8319 +/- 1.46554  | 22.087 +/- 0.552932
> >       | 389.367x             | 402.499x
> > ------+----------------------+------------------------
> > run 2 | 22.4286 +/- 0.726449 | 22.216 +/- 0.916222
> >       | 396.369x             | 400.162x
> > ------+----------------------+------------------------
> > run 2 | 22.5688 +/- 0.831341 | 22.4902 +/- 1.04881
> >       | 393.906x             | 395.283x
> >
> > Bug: webrtc:10480
> > Change-Id: I1fd54077a32e25e46196c8e18f003cd0ffd503e1
> > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/191703
> > Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
> > Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
> > Cr-Commit-Position: refs/heads/master@{#32572}
>
> TBR=alessiob@webrtc.org,kwiberg@webrtc.org
>
> Change-Id: I57a8f937ade0a35e1ccf0e229c391cc3a10e7c48
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: webrtc:10480
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/192621
> Reviewed-by: Alessio Bazzica <alessiob@webrtc.org>
> Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#32578}

TBR=alessiob@webrtc.org,kwiberg@webrtc.org

# Not skipping CQ checks because this is a reland.

Bug: webrtc:10480
Change-Id: I1d510697236255d8c0cca405e90781f5d8c6a3e6
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/192783
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Reviewed-by: Alessio Bazzica <alessiob@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32587}
This commit is contained in:
Alessio Bazzica
2020-11-11 12:54:39 +01:00
committed by Commit Bot
parent 5a37b94029
commit f2a2fe84b8
5 changed files with 27 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

35
modules/audio_processing/agc2/rnn_vad/pitch_search_internal.cc
View File

@ -172,7 +172,7 @@ int ComputePitchPeriod24kHz(
// Auto-correlation energy normalized by frame energy.
const float numerator =
auto_correlation[inverted_lag] * auto_correlation[inverted_lag];
const float denominator = y_energy[kMaxPitch24kHz - inverted_lag];
const float denominator = y_energy[inverted_lag];
// Compare numerator/denominator ratios without using divisions.
if (numerator * best_denominator > best_numerator * denominator) {
best_inverted_lag = inverted_lag;
@ -256,19 +256,19 @@ void Decimate2x(rtc::ArrayView<const float, kBufSize24kHz> src,
void ComputeSlidingFrameSquareEnergies24kHz(
rtc::ArrayView<const float, kBufSize24kHz> pitch_buffer,
rtc::ArrayView<float, kRefineNumLags24kHz> yy_values) {
float yy = ComputeAutoCorrelation(kMaxPitch24kHz, pitch_buffer);
yy_values[0] = yy;
static_assert(kMaxPitch24kHz - (kRefineNumLags24kHz - 1) >= 0, "");
rtc::ArrayView<float, kRefineNumLags24kHz> y_energy) {
float yy = std::inner_product(pitch_buffer.begin(),
pitch_buffer.begin() + kFrameSize20ms24kHz,
pitch_buffer.begin(), 0.f);
y_energy[0] = yy;
static_assert(kMaxPitch24kHz - 1 + kFrameSize20ms24kHz < kBufSize24kHz, "");
for (int lag = 1; lag < kRefineNumLags24kHz; ++lag) {
const int inverted_lag = kMaxPitch24kHz - lag;
const float y_old = pitch_buffer[inverted_lag + kFrameSize20ms24kHz];
const float y_new = pitch_buffer[inverted_lag];
yy -= y_old * y_old;
yy += y_new * y_new;
yy = std::max(0.f, yy);
yy_values[lag] = yy;
static_assert(kMaxPitch24kHz < kRefineNumLags24kHz, "");
for (int inverted_lag = 0; inverted_lag < kMaxPitch24kHz; ++inverted_lag) {
yy -= pitch_buffer[inverted_lag] * pitch_buffer[inverted_lag];
yy += pitch_buffer[inverted_lag + kFrameSize20ms24kHz] *
pitch_buffer[inverted_lag + kFrameSize20ms24kHz];
yy = std::max(1.f, yy);
y_energy[inverted_lag + 1] = yy;
}
}
@ -382,7 +382,7 @@ PitchInfo ComputeExtendedPitchPeriod48kHz(
float y_energy; // Energy of the sliding frame `y`.
};
const float x_energy = y_energy[0];
const float x_energy = y_energy[kMaxPitch24kHz];
const auto pitch_strength = [x_energy](float xy, float y_energy) {
RTC_DCHECK_GE(x_energy * y_energy, 0.f);
return xy / std::sqrt(1.f + x_energy * y_energy);
@ -394,7 +394,7 @@ PitchInfo ComputeExtendedPitchPeriod48kHz(
std::min(initial_pitch_period_48kHz / 2, kMaxPitch24kHz - 1);
best_pitch.xy =
ComputeAutoCorrelation(kMaxPitch24kHz - best_pitch.period, pitch_buffer);
best_pitch.y_energy = y_energy[best_pitch.period];
best_pitch.y_energy = y_energy[kMaxPitch24kHz - best_pitch.period];
best_pitch.strength = pitch_strength(best_pitch.xy, best_pitch.y_energy);
// Keep a copy of the initial pitch candidate.
const PitchInfo initial_pitch{best_pitch.period, best_pitch.strength};
@ -435,8 +435,9 @@ PitchInfo ComputeExtendedPitchPeriod48kHz(
const float xy_secondary_period = ComputeAutoCorrelation(
kMaxPitch24kHz - dual_alternative_period, pitch_buffer);
const float xy = 0.5f * (xy_primary_period + xy_secondary_period);
const float yy = 0.5f * (y_energy[alternative_pitch.period] +
y_energy[dual_alternative_period]);
const float yy =
0.5f * (y_energy[kMaxPitch24kHz - alternative_pitch.period] +
y_energy[kMaxPitch24kHz - dual_alternative_period]);
alternative_pitch.strength = pitch_strength(xy, yy);
// Maybe update best period.

4
modules/audio_processing/agc2/rnn_vad/pitch_search_internal.h
View File

@ -62,10 +62,10 @@ void Decimate2x(rtc::ArrayView<const float, kBufSize24kHz> src,
// corresponding pitch period.
// Computes the sum of squared samples for every sliding frame `y` in the pitch
// buffer. The indexes of `yy_values` are lags.
// buffer. The indexes of `y_energy` are inverted lags.
void ComputeSlidingFrameSquareEnergies24kHz(
rtc::ArrayView<const float, kBufSize24kHz> pitch_buffer,
rtc::ArrayView<float, kRefineNumLags24kHz> yy_values);
rtc::ArrayView<float, kRefineNumLags24kHz> y_energy);
// Top-2 pitch period candidates. Unit: number of samples - i.e., inverted lags.
struct CandidatePitchPeriods {

2
modules/audio_processing/agc2/rnn_vad/pitch_search_internal_unittest.cc
View File

@ -42,7 +42,7 @@ TEST(RnnVadTest, ComputeSlidingFrameSquareEnergies24kHzWithinTolerance) {
computed_output);
auto square_energies_view = test_data.GetPitchBufSquareEnergiesView();
ExpectNearAbsolute({square_energies_view.data(), square_energies_view.size()},
computed_output, 3e-2f);
computed_output, 1e-3f);
}
// Checks that the estimated pitch period is bit-exact given test input data.

2
modules/audio_processing/agc2/rnn_vad/pitch_search_unittest.cc
View File

@ -42,7 +42,7 @@ TEST(RnnVadTest, PitchSearchWithinTolerance) {
pitch_estimator.Estimate({lp_residual.data(), kBufSize24kHz});
EXPECT_EQ(expected_pitch_period, pitch_period);
EXPECT_NEAR(expected_pitch_strength,
pitch_estimator.GetLastPitchStrengthForTesting(), 1e-5f);
pitch_estimator.GetLastPitchStrengthForTesting(), 15e-6f);
}
}
}

5
modules/audio_processing/agc2/rnn_vad/test_utils.cc
View File

@ -10,6 +10,7 @@
#include "modules/audio_processing/agc2/rnn_vad/test_utils.h"
#include <algorithm>
#include <memory>
#include "rtc_base/checks.h"
@ -86,6 +87,10 @@ PitchTestData::PitchTestData() {
ResourcePath("audio_processing/agc2/rnn_vad/pitch_search_int", "dat"),
1396);
test_data_reader.ReadChunk(test_data_);
// Reverse the order of the squared energy values.
// Required after the WebRTC CL 191703 which switched to forward computation.
std::reverse(test_data_.begin() + kBufSize24kHz,
test_data_.begin() + kBufSize24kHz + kNumPitchBufSquareEnergies);
}
PitchTestData::~PitchTestData() = default;