This CL corrects the EchoAudibility and StationarityEstimator
code to work properly with multiple channels.
It also changes the naming of the FilterDelayBlocks() method
to better reflect what it does.
The changes have been verified to be bitexact over a large number
of recordings.
Bug: webrtc:10913
Change-Id: I070b531efcdff4c33f70fd5b37fbb556dcebe5b4
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/156565
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29482}
This CL adds support for multiple channels in the reverb
modelling. As a side effect, it also partly adds multi-channel
supports for the sections of the code.
Beyond adding the multi-channel support, a bug is fixed as part of
this CL. Since the bug fix affects the bitexactness, as a safety
precaution the CL includes the ability to override the bugfix.
Apart from the contributions from the bugfix, the changes have
been verified to be bitexact for a large set of mono recordings.
Bug: webrtc:10913
Change-Id: I1f307b532be85ef4182f8db41384f44d40a25219
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/156382
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29456}
This change keeps the original 48 kHz signal and uses it for the
fullband processing given that the following requirements are
fulfilled:
- Input signal is 48 kHz
- Output signal is 48 kHz
- Multiband processing is performed at 32 kHz
- The multiband processing does not modify the original signal
This avoids unnecessary, lossy resampling and band merging.
Bug: b/130016532
Change-Id: I690c26faba07eab0cbff6c0a95a81d89255dd1a1
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/155966
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29425}
This CL passes the spectral power estimates for all channels into
the AecState.
Bug: webrtc:10913
Change-Id: Ie3b5c443be0c63f205e23ed2bfea06d9c447eb39
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/156165
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29417}
This CL removes the redundant class in preparation
for adding multichannel functionality to the
reverb computation.
The changes are bitexact.
Bug: webrtc:10913
Change-Id: I284665f7143cb5e1c79bfa573638fdff5f2411c9
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/155960
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29414}
Multi-channel behaviors introduced in this CL:
- All filters are analyzed independently. The filtering is considered
consistent if any filter is consistent.
- The filter echo path gain used to detect saturation is maxed across
capture channels.
- The filter delay is taken to be the minimum of all filters:
Any module that looks in the render data starting from the filter
delay will iterate over all render audio present in any channel.
- The FilterAnalyzer will consider a render block to be active if any
render channel has activity.
The changes in the CL has been shown to be bitexact on a
large set of mono recordings.
Bug: webrtc:10913
Change-Id: I1e360cd7136ee82d1f6e0f8a1459806e83f4426d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/155363
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29408}
This CL adds support for multichannel in the residual echo
estimator code. It also adds placeholder functionality in
the surrounding code to ensure that the residual echo
estimator receives the require inputs.
The changes in the CL has been shown to be bitexact on a
large set of mono recordings.
Bug: webrtc:10913
Change-Id: I726128ca928648b1dcf36c5f479eb243f3ff3f96
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/155361
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29400}
This CL adds support in the echo subtractor for handling multiple
capture and render channels.
The changes have passed bitexactness tests for substantial set
of mono recordings.
Bug: webrtc:10913
Change-Id: Ib448c9edf172ebc31e8c28db7b2f2a389a53adb9
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/155168
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29389}
This CL contains the first step of adding multi-channel support to the
echo subtractor.
The CL is bitexact for the mono case.
Bug: webrtc:10913
Change-Id: I10647b45c692bc001407afc6ff00e26a3e2cffaa
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/154356
Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
Commit-Queue: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29303}
Simplifying the use of signal transition and removing unused code.
Bug: webrtc:8671
Change-Id: I0b845405727936b2fa7df7c92ad2e83bea3bc823
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/154348
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29298}
Suppression filter is extended to support the synthesis
of multiple channels. This CL is also a major clean-up of ApplyGain.
The CL has been tested for bit-exactness for single channel output.
Bug: webrtc:10913
Change-Id: I1319f127981552e17dec66701a248d34dcf0e563
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/154341
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29284}
Echo remover processes all microphone signals. Suppression gains are
computed separately for each capture signal. The minimum gains determine
the final suppression gains applied.
Only the first channel is synthesized. A follow-up CL will add the
synthesis of the remaining channels.
Bug: webrtc:10913
Change-Id: Ife7e74c9a9c6c208fca3992e3cfa840b6b7afcfa
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/153526
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29269}
Currently, APM fakes multichannel in two ways:
- With injected AECs, capture processing is only performed on the left
channel. The result is copied into the other channels.
- With multichannel render audio, all channels are mixed into one
before analysing.
This CL adds a flag to disable these behaviors, ensuring proper
multichannel processing happens throughout the APM pipeline.
Adds killswitches to separately disable render / capture multichannel.
Additionally - AEC3 currently crashes when running with multichannel.
This CL adds the missing pieces to at least have it run without
triggering any DCHECKS, including making the high pass filter properly
handle multichannel.
Bug: webrtc:10913, webrtc:10907
Change-Id: I38795bf8f312b959fcc816a056fba2c68d4e424d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/152483
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29248}
In this CL:
- Render signal analyzer considers a frequency bin a narrow band
(peak) if any channel exhibits narrowband (-peak) behavior.
- The unit tests have to fill frames with noise because small
inaccuracies in the FFT spectrum lead to consistent "narrow bands"
despite spectrum being essentially flat.
Bug: webrtc:10913
Change-Id: I8fa181412c0ee1beeacfda37ffef18251d5f0cd7
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151912
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29176}
This CL introduces the handling of multiple microphone channels in
the EchoRemover layer.
The implementation is done such as to support an arbitrary number of
channels in a way that balances stack and heap-space usage.
Bug: webrtc:10913
Change-Id: I475369de6c463b8fe2d7e53799d7322eefb6938f
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151647
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29140}
Multichannel signals are downmixed to mono before decimation and
delay estimation. This is useful when not all channels play
audio content. The feature can be toggled in the AEC3 configuration.
Bug: webrtc:10913
Change-Id: I7d40edf7732bb51fec69e7f3ca063d821c5069c4
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151762
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29126}
This CL reduces the complexity and heap usage of the adaptive filter
in AEC3 by avoiding to compute these for the shadow
filter. In particular it
-Moves to compute the ERL, frequency response and impulse response
on an on-demand basis.
-Stores the ERL, frequency response and impulse response outside
of the adaptive filter.
All the changes have been tested for bitexactness on a sizeable
amount of recordings.
Bug: webrtc:10913
Change-Id: If83c236a6e3f2e489be129b9ebf6143a72f521d1
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151138
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29081}
All channels are populated by RenderDelayBuffer. but all other
dependent modules are hardcoded to do their regular mono processing
on the first channel.
Bug: webrtc:10913
Tested: Bitexactness on a large set of aecdumps
Change-Id: I11d11aa0ad3da0f244c0ec020d2c9f0f4a735834
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151640
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29079}
The VectorBuffer and MatrixBuffer names are too generic for their use case.
Bug: webrtc:10913
Change-Id: Ideecd0d27e07487a85a61dc6474e69733d07dcd6
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151602
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29076}
All dependent modules are hardcoded to do their regular mono processing on the first channel.
This _almost_ makes RenderBuffer multi-channel: FftData is still only mono.
Bug: webrtc:10913
Change-Id: Id5cc34dbabfe59e1cc72a9675dc7979794e870ed
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151139
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29074}
This CL propagates the number of render and capture channels into
the echo subtractor and the adaptive filters.
Bug: webrtc:10913
Change-Id: I5ffff24ff64b7cc0f262bf008b34b6dfca1e78f9
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151300
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29059}
This is a reland of a66395e72f
Original change's description:
> Reland "Add core multi-channel pipeline in AEC3 This CL adds basic the basic pipeline to support multi-channel processing in AEC3."
>
> This is a reland of f3a197e553
>
> Original change's description:
> > Add core multi-channel pipeline in AEC3
> > This CL adds basic the basic pipeline to support multi-channel
> > processing in AEC3.
> >
> > Apart from that, it removes the 8 kHz processing support in several
> > places of the AEC3 code.
> >
> > Bug: webrtc:10913
> > Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137
> > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332
> > Commit-Queue: Per Åhgren <peah@webrtc.org>
> > Reviewed-by: Sam Zackrisson <saza@webrtc.org>
> > Cr-Commit-Position: refs/heads/master@{#29017}
>
> Bug: webrtc:10913
> Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124
> Reviewed-by: Sam Zackrisson <saza@webrtc.org>
> Commit-Queue: Per Åhgren <peah@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#29034}
Bug: webrtc:10913
Change-Id: Id8da5666df8c86f290c73ad5dc9958199f1a7ebe
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151127
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29042}
This is a reland of f3a197e553
Original change's description:
> Add core multi-channel pipeline in AEC3
> This CL adds basic the basic pipeline to support multi-channel
> processing in AEC3.
>
> Apart from that, it removes the 8 kHz processing support in several
> places of the AEC3 code.
>
> Bug: webrtc:10913
> Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332
> Commit-Queue: Per Åhgren <peah@webrtc.org>
> Reviewed-by: Sam Zackrisson <saza@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#29017}
Bug: webrtc:10913
Change-Id: Ifc4b13bd994cfd22dca8f8755fa5700617cc379d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/151124
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29034}
This reverts commit f3a197e553.
Reason for revert: Speculative revert, as this may'be broken some build bots
Original change's description:
> Add core multi-channel pipeline in AEC3
> This CL adds basic the basic pipeline to support multi-channel
> processing in AEC3.
>
> Apart from that, it removes the 8 kHz processing support in several
> places of the AEC3 code.
>
> Bug: webrtc:10913
> Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332
> Commit-Queue: Per Åhgren <peah@webrtc.org>
> Reviewed-by: Sam Zackrisson <saza@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#29017}
TBR=saza@webrtc.org,peah@webrtc.org
Change-Id: I877d2993b9ccf024bd1d57bca1513c3e24d0bed3
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:10913
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150940
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29022}
This CL adds basic the basic pipeline to support multi-channel
processing in AEC3.
Apart from that, it removes the 8 kHz processing support in several
places of the AEC3 code.
Bug: webrtc:10913
Change-Id: If5b75fa325ed0071deea94a7546cb4a7adf22137
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150332
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#29017}
This CL removes and replaces the legacy fixed-point high-pass filter in
APM with the floating point high-pass filter in AEC3.
Bug: webrtc:10907
Change-Id: I88cf8f622ab139e4ffa97f89a72425aa3becfc58
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150103
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28950}
This is a reland of 81c0cf287c
Original change's description:
> Simplification and refactoring of the AudioBuffer code
>
> This CL performs a major refactoring and simplification
> of the AudioBuffer code that.
> -Removes 7 of the 9 internal buffers of the AudioBuffer.
> -Avoids the implicit copying required to keep the
> internal buffers in sync.
> -Removes all code relating to handling of fixed-point
> sample data in the AudioBuffer.
> -Changes the naming of the class methods to reflect
> that only floating point is handled.
> -Corrects some bugs in the code.
> -Extends the handling of internal downmixing to be
> more generic.
>
> Bug: webrtc:10882
> Change-Id: I12c8af156fbe366b154744a0a1b3d926bf7be572
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/149828
> Commit-Queue: Per Åhgren <peah@webrtc.org>
> Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#28928}
Bug: webrtc:10882
Change-Id: I2ddf327e80a03468c41662ae63c619ff34f2363a
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150101
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28938}
This reverts commit 81c0cf287c.
Reason for revert: internal test failures
Original change's description:
> Simplification and refactoring of the AudioBuffer code
>
> This CL performs a major refactoring and simplification
> of the AudioBuffer code that.
> -Removes 7 of the 9 internal buffers of the AudioBuffer.
> -Avoids the implicit copying required to keep the
> internal buffers in sync.
> -Removes all code relating to handling of fixed-point
> sample data in the AudioBuffer.
> -Changes the naming of the class methods to reflect
> that only floating point is handled.
> -Corrects some bugs in the code.
> -Extends the handling of internal downmixing to be
> more generic.
>
> Bug: webrtc:10882
> Change-Id: I12c8af156fbe366b154744a0a1b3d926bf7be572
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/149828
> Commit-Queue: Per Åhgren <peah@webrtc.org>
> Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#28928}
TBR=gustaf@webrtc.org,peah@webrtc.org
Change-Id: I2729e3ad24b3a9b40b368b84cb565c859e79b51e
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:10882
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/150084
Reviewed-by: Steve Anton <steveanton@webrtc.org>
Commit-Queue: Steve Anton <steveanton@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28931}
This CL performs a major refactoring and simplification
of the AudioBuffer code that.
-Removes 7 of the 9 internal buffers of the AudioBuffer.
-Avoids the implicit copying required to keep the
internal buffers in sync.
-Removes all code relating to handling of fixed-point
sample data in the AudioBuffer.
-Changes the naming of the class methods to reflect
that only floating point is handled.
-Corrects some bugs in the code.
-Extends the handling of internal downmixing to be
more generic.
Bug: webrtc:10882
Change-Id: I12c8af156fbe366b154744a0a1b3d926bf7be572
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/149828
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28928}
This change fixes a bug where the initial delay could be set incorrectly.
Bug: webrtc:10896
Change-Id: I66b2234b69c46639488f4561e973384001230861
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/149820
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28894}
This change reduces the level of several non-critical log messages in
order to reduce log spamming.
Bug: webrtc:8671
Change-Id: I6faae7a2ae4eeafd18c2770208485a75ad946e20
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/148528
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28809}
This CL removes a long unused fallback behavior for the reverb
computation.
Bug: webrtc:8671
Change-Id: I4b57795a9bb33769237858f40392881ee235653e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/148520
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28802}
As non-linear mode uses a suppressed version of y (not e) as output, this change
uses Y2, rather than E2, as nearend spectrum when computing the suppression gains.
E2 is still used in linear mode.
This change also affects how the minimum suppression gains are calculated. The
minimum gain is now min_echo_power / weighted_residual_echo.
Bug: webrtc:10550
Change-Id: I2904c5a09dd64b06bf25eb5a37c18dab50297794
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/133023
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27629}
