Sample rates not divisible by 100, in particular 11025 Hz and 22050 Hz, have long been used with APM in Chrome, but the support has never been stated explicitly.
This CL makes minor modifications to the APM API to clarify how rates are handled when 10 ms is not an integer number of samples. Unit tests are also extended to cover this case better.
This does not update all references to 10 ms and implicit floor(sample_rate/100) computations, but it does at least take us closer to a correct API.
Note that not all code needs to support these sample rates. For example, audio processing submodules only need to operate on the native APM rates 16000, 32000, 48000.
Bug: chromium:1332484
Change-Id: I1dad15468f6ccb9c0d4d09c5819fe87f8388d5b8
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/268769
Reviewed-by: Henrik Andreassson <henrika@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Reviewed-by: Ivo Creusen <ivoc@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37682}
This is the first step of migrating
AudioProcessing::CreateAndAttachAecDump() from using std::string to
absl::string_view.
Bug: webrtc:13579
Change-Id: I8fc373e7ac55fd8e96bb0b01d1a30e28883ac9a2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/269400
Commit-Queue: Ali Tofigh <alito@webrtc.org>
Reviewed-by: Ivo Creusen <ivoc@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37631}
It is now easier to fully test `AgcManagerDirect` with different values
for the used field trials. In particular, this CL adds tests for the
field trial named `WebRTC-Audio-2ndAgcMinMicLevelExperiment`.
1. `UnmutingRaisesTooLowVolume` and `MicVolumeIsLimited`
The expectations for the lowest input volume are not hard-coded anymore
since the parametrized tests use different values for the enforced
minimum.
2. `RecoveryAfterManualLevelChangeBelowMin`
The recovery behavior after manual input volume change depends on
whether the minimum input volume is overridden. When that's the case,
the minimum volume is applied immediately after the manual adjustment.
Hence, the existing test is left and a parametrized version of it has been added to test the "instant recovery" behavior. The latter test is
skipped when the minimum input volume is not overridden since that case
is covered by the existing test.
Bug: chromium:1275566
Change-Id: Ib0d4427b32b88f33138d4062b365916a3c47a406
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/268900
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Reviewed-by: Hanna Silen <silen@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37577}
Stop using TEST_F; that will make it easier to switch to parametric
tests that are needed to correctly test `AgcManagerDirect`.
"Avoid fixtures where reasonable."
Source: https://abseil.io/tips/122
Bug: chromium:1275566
Change-Id: I2d73a0913eb2349144f63bd17ab4d6efa245e472
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/268766
Reviewed-by: Hanna Silen <silen@webrtc.org>
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37556}
This reverts commit d0a6fd239c.
Reason for revert: reland the bug fix
Original change's description:
> Revert "`AgcManagerDirect`: stop enforcing min mic level override with 0 level"
>
> This reverts commit e76daab8b3.
>
> Reason for revert: revert required to revert the parent CL
>
> Original change's description:
> > `AgcManagerDirect`: stop enforcing min mic level override with 0 level
> >
> > https://webrtc-review.googlesource.com/c/src/+/250141 introduced a bug
> > due to which the min mic level override is always enforced, if specified
> > even if the user manually adjusts the mic level to zero.
> >
> > This CL fixes that bug, the changes run behind a kill switch.
> >
> > TESTED=Test video call on Chromium on Mac; input volume not adjusted after zeroing it from the system preferences UI
> >
> > Bug: chromium:1275566
> > Change-Id: I18ce2e5970d3002b301f51f84544583c64982d57
> > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/267844
> > Reviewed-by: Hanna Silen <silen@webrtc.org>
> > Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
> > Cr-Commit-Position: refs/heads/main@{#37460}
>
> Bug: chromium:1275566
> Change-Id: I6d22d8f3fafdc7da3814827b9b69146a506595db
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/268468
> Bot-Commit: rubber-stamper@appspot.gserviceaccount.com <rubber-stamper@appspot.gserviceaccount.com>
> Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
> Cr-Commit-Position: refs/heads/main@{#37515}
Bug: chromium:1275566
Change-Id: I7198587dec2a153270e8beb714e9dacccdaae806
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/268544
Reviewed-by: Hanna Silen <silen@webrtc.org>
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37530}
This reverts commit c9cad23274.
Reason for revert: add back field trial
Original change's description:
> Min mic analog level: override minimum and behavior on Mac
>
> This CL removes the `WebRTC-Audio-AgcMinMicLevelExperiment` field trial
> and always enables the code path behind that flag on Mac. In summary,
> the analog AGC behaves as follows on Mac:
> 1. the minimum level is overridden to 20
> 2. the minimum is applied even when clipping is detected
> 3. when the level is manually adjusted to 0, the minimum level is
> enforced - i.e., 20
>
> Note that the 3rd property had been unintentionally added when the
> changes were added behind the aforementioned field trial. This will
> be fixed in a follow-up CL.
>
> Bug: chromium:1275566
> Change-Id: If184c4455a0780fcd94f55141af34460c152e3c3
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/266488
> Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
> Reviewed-by: Hanna Silen <silen@webrtc.org>
> Cr-Commit-Position: refs/heads/main@{#37459}
Bug: chromium:1275566
Change-Id: I00a37ad9e16efc49f721558d25af16efd5f3db8c
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/268540
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Reviewed-by: Hanna Silen <silen@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37521}
This reverts commit e76daab8b3.
Reason for revert: revert required to revert the parent CL
Original change's description:
> `AgcManagerDirect`: stop enforcing min mic level override with 0 level
>
> https://webrtc-review.googlesource.com/c/src/+/250141 introduced a bug
> due to which the min mic level override is always enforced, if specified
> even if the user manually adjusts the mic level to zero.
>
> This CL fixes that bug, the changes run behind a kill switch.
>
> TESTED=Test video call on Chromium on Mac; input volume not adjusted after zeroing it from the system preferences UI
>
> Bug: chromium:1275566
> Change-Id: I18ce2e5970d3002b301f51f84544583c64982d57
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/267844
> Reviewed-by: Hanna Silen <silen@webrtc.org>
> Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
> Cr-Commit-Position: refs/heads/main@{#37460}
Bug: chromium:1275566
Change-Id: I6d22d8f3fafdc7da3814827b9b69146a506595db
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/268468
Bot-Commit: rubber-stamper@appspot.gserviceaccount.com <rubber-stamper@appspot.gserviceaccount.com>
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37515}
https://webrtc-review.googlesource.com/c/src/+/250141 introduced a bug
due to which the min mic level override is always enforced, if specified
even if the user manually adjusts the mic level to zero.
This CL fixes that bug, the changes run behind a kill switch.
TESTED=Test video call on Chromium on Mac; input volume not adjusted after zeroing it from the system preferences UI
Bug: chromium:1275566
Change-Id: I18ce2e5970d3002b301f51f84544583c64982d57
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/267844
Reviewed-by: Hanna Silen <silen@webrtc.org>
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37460}
This CL removes the `WebRTC-Audio-AgcMinMicLevelExperiment` field trial
and always enables the code path behind that flag on Mac. In summary,
the analog AGC behaves as follows on Mac:
1. the minimum level is overridden to 20
2. the minimum is applied even when clipping is detected
3. when the level is manually adjusted to 0, the minimum level is
enforced - i.e., 20
Note that the 3rd property had been unintentionally added when the
changes were added behind the aforementioned field trial. This will
be fixed in a follow-up CL.
Bug: chromium:1275566
Change-Id: If184c4455a0780fcd94f55141af34460c152e3c3
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/266488
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Reviewed-by: Hanna Silen <silen@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37459}
This CL duplicates a few lines of utility code from
//modules/audio_processing:audioproc_test_utils (which contains more
testonly things) and allows the possibility to remove testonly from
the unpack_aecdump tool.
Bug: b/237526033
Change-Id: If2e1dd4cc825429c496091cf8640c67069fb6e6f
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/267701
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37437}
Look for first echo (and not only the strongest one) on the same matched
filter.
This change is bit exact with previous version when `pre_echo` is false.
Author: Jesús de Vicente Peña <devicentepena@webrtc.org>
Bug: webrtc:14205
Change-Id: I6782eaa1d690b0df78d00f6d425a85c951b2ca9d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/266321
Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
Commit-Queue: Lionel Koenig <lionelk@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37360}
When the `WebRTC-Audio-TransientSuppressorVadMode-RnnVad` field trial
is set, APM now uses (i) its RNN VAD sub-module to compute the voice
probability, (ii) that probability for TS and (iii) a temporally
delayed version of it for AGC2 (the delay introduced by TS is taken
into account).
Bug: webrtc:13663
Change-Id: Ic0f245c3f00d318c19bb01d3dbc2d5176c90f851
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/266362
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Reviewed-by: Hanna Silen <silen@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37291}
Add a VoiceActivityDetectorWrapper submodule in AudioProcessingImpl
and enable injecting speech probability into GainController2.
Bug: webrtc:13663
Change-Id: I05e13b737d085b45ac8ce76660191867c56834c2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/265166
Commit-Queue: Hanna Silen <silen@webrtc.org>
Reviewed-by: Alessio Bazzica <alessiob@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#37275}
Puts the whole block in contiguous memory and reduce pointer look-up.
The change has been verified to be bit-exact.
Bug: webrtc:14089
Change-Id: I264aaf764bf53a29f23249105f704b2fdbd7e51c
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/263203
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36983}
The high-band gain is corrected by fixing the computation of the
low-band energy
Bug: webrtc:14108
Change-Id: I5033287de57aedcd91bb71623ca2862519ffb35b
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/263201
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Reviewed-by: Jesus de Vicente Pena <devicentepena@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36972}
This change adds a Block class to reduce the need for std::vector<std::vector<std::vector<float>>>. This make the code
easier to read and less error prone.
It also enables future changes to the underlying data structure of a
block. For instance, the data of all bands and channels could be stored
in a single vector.
The change has been verified to be bit-exact.
Bug: webrtc:14089
Change-Id: Ied9a78124c0bbafe0e912017aef91f7c311de2ae
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/262252
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36968}
To quote rfc6464:
The audio level for digital silence -- for a muted audio source, for
example -- MUST be represented as 127 (-127 dBov), regardless of the
dynamic range of the encoded audio format.
The behavior in webrtc is correct that digital silence is represented
with 127, but it is also possible to get a value of 127 for not quite
digitally silent audio buffer (as in, not strictly 0s).
Bug: webrtc:14029
Change-Id: I7ff8698a7e4d5c0960c667fd1cc961838e269456
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/261244
Commit-Queue: Tomas Gunnarsson <tommi@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36793}
* Structs with user-declared constructors are no longer considered
aggregates, so remove the declarations when possible
* Types of both arguments to "==" must match to avoid "ambiguous
function call" warning
* Various types of math involving enums are deprecated, so replace with
constexprs where necessary
* ABSL_CONST_INIT must be used on definition as well as declaration
* volatile memory may no longer be read from and written to by the same
operator, so replace e.g. "n++" with "n = n + 1"
* Replace an outdated check for no_unique_address support with
__has_cpp_attribute
* std::result_of(f(x)) has been removed, replace with
std::invoke_result(f, x)
Bug: chromium:1284275
Change-Id: I77b366ab1da7eb2c1e4c825b2714417c31ee5903
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/261221
Auto-Submit: Peter Kasting <pkasting@chromium.org>
Reviewed-by: Tomas Gunnarsson <tommi@google.com>
Reviewed-by: Tomas Gunnarsson <tommi@webrtc.org>
Commit-Queue: Tomas Gunnarsson <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36786}
While the target has a restricted visibility, since it was in rtc_base_approved
public deps, a lot of targets were able to bypass the visibility check.
So we remove the visibility restrictions and use the dependency explicitely
everywhere instead.
Bug: webrtc:8603
Change-Id: I94a03fdf7f94c54ab72081a58dd648e2cca73d17
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/258944
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Commit-Queue: Florent Castelli <orphis@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36566}
Adds two metrics for stereo detection:
- An enum indicating whether the last 10 seconds contained persistent stereo content or not, logged every 10 seconds.
- An enum indicating whether any persistent stereo content at all has been detected, logged at the end of the AEC lifetime.
These metrics allow us to assess:
- What proportion of all audio is treated as stereo.
- What proportion of sessions encounter any significant stereo content. If this is unexpectedly high, the stereo detection code may need fine tuning.
Metrics are only logged for component lifetimes exceeding 5 seconds: This is to avoid brief AEC lifetimes due to internal resets etc within APM.
Corresponding Chrome CL for XML histogram declarations:
https://crrev.com/c/3579317
Bug: chromium:1295710
Change-Id: I93e2bf74588cf4bb2a8922dbfad079bccab01456
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/258760
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36537}
During temporary stereo content when the AEC3 uses a mono reference signal, the signal is downmixed by averaging instead of using only the left channel.
Additionally, temporary stereo content is flagged as an echo path change.
Tested: Modified local build: Verified stereo mode entered / left in accordance with hysteresis and timeout thresholds. Verified temporary stereo detected during temporary stereo playout. Made an aecdump and inspected content.
Bug: chromium:1295710
Change-Id: I6bd53e615dfb3ec39bc1c73275b7d6d599ac7c57
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/258481
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36504}
Even if playout audio is only very briefly stereo, the AEC will enter stereo processing mode. To save CPU and improve AEC performance, this CL adds a hysteresis period before treating playout as stereo.
The feature is enabled by default in the AEC3 config.
Bug: chromium:1295710
Change-Id: I29116ab2e7823e25a02aa3b66a1c619f1d966d9e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/258479
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36503}
If playout audio is temporarily stereo, the AEC will currently enter stereo processing mode indefinitely. To save CPU and improve AEC performance, this CL adds support for falling back to mono after a period of no stereo.
The feature is enabled by default in the AEC3 config.
Bug: chromium:1295710
Change-Id: I690b5b22f8407f950bf41f3bcaa9ca0138452157
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/258421
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36502}
This CL adds a component in the TS implementation to return a delayed
version of the voice probability values observed when `Suppress()` is
called. That is needed in order to temporally align the voice
probability values to the processed audio since TS adds algorithmic
delay.
Bug: webrtc:13663
Change-Id: I5041ace3939d2ce7ba084ae703428e66f1aa06be
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/255860
Reviewed-by: Hanna Silen <silen@webrtc.org>
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36496}
The values returned by `TransientSuppressor::Initialize()` and by
`TransientSuppressor::Suppress()` are never used.
Bug: webrtc:13663
Change-Id: I20b8afb5a66f49e5ebaf132acf8bcd1c4292a5f8
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/255822
Reviewed-by: Jesus de Vicente Pena <devicentepena@webrtc.org>
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36492}
More robust API option that allows to fully initialize TS when created.
Bug: webrtc:13663
Change-Id: I42c38612ef772eb6d0bbde49d04ea39332a0e3c7
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/255821
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36490}
Adding a delay unit to be used in the APM Transient Suppressor (TS)
sub-module through which the observerd voice probabilities are
temporally aligned to the audio processed by TS, which introduces
algorithmic delay.
Bug: webrtc:13663
Change-Id: I2136c303914580851c742d8db89478a13b06dacb
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/255680
Reviewed-by: Hanna Silen <silen@webrtc.org>
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36487}
It is now required to specify which VAD is used to compute the speech
probability passed when `TransientSuppressor::Suppress()` is called.
In this way, it is possible to adapt parameters and/or logic of a
`TransientSuppressor` implementation to the behavior of the used
VAD. This CL also adds a "no VAD" mode option, which ignores the speech
probability argument passed when `Suppress()` and always applies mild
suppression to preserve transparency.
Finally, this CL adds a field trial to choose which VAD is used by
APM for transient suppression. Wiring the RNN VAD to TS will be done
in a follow-up CL.
Bug: webrtc:13663
Change-Id: I21ed49f91875a4ee0f04db97ea87c0dbc3db7f8a
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/250962
Reviewed-by: Hanna Silen <silen@webrtc.org>
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36485}
The features have two safety fallbacks:
- multichannel config has a killswitch WebRTC-Aec3SetupSpecificDefaultConfigDefaultsKillSwitch
- stereo detection has a killswitch WebRTC-Aec3StereoContentDetectionKillSwitch
Both features are enabled by default in the AEC3 config.
Tested: Bitexact on a large number of aecdumps.
Bug: chromium:1295710
Change-Id: I340cdc9140dacd4ca22d0911eb9f732b6cf8b226
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/258129
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36482}
Apart from making the construction more straightforward, this change allows recreating the BlockProcessor at runtime. This is used to change parameterization at runtime in an upcoming CL [1].
[1] https://webrtc-review.googlesource.com/c/src/+/258129
Tested: Bitexact on a large number of aecdumps.
Bug: chromium:1295710
Change-Id: I2e0275c5c97044cb4370042633266b193c06b960
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/258100
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36473}
