Currently, FFT is performance when AEC buffers farend signal. This has some drawbacks
1. memory inefficiency: two ring buffers are needed;
2. computation inefficiency: if ringbuffer gets wrapped around, some FFT computation will be wasted;
3. accessibility: the main AEC function looses accessibility to the time-domain signal.
Therefore, this CL tries to buffer time domain data, which is buffered any way if a debugging macro is defined, and calculate the FFTs where they are actually used.
BUG=
Review URL: https://codereview.webrtc.org/1512573003
Cr-Commit-Position: refs/heads/master@{#11091}
-Made the component error messages generic to be an unspecified error message.
BUG=webrtc:5099
Review URL: https://codereview.webrtc.org/1404743003
Cr-Commit-Position: refs/heads/master@{#10570}
We use this Config struct for enabling/disabling the delay agnostic
AEC. This change renames it to DelayAgnostic for readability reasons.
NOTE: The logic is reversed in this CL. The old ReportedDelay config
turned DA-AEC off, while the new DelayAgnostic turns it on.
The old Config is kept in parallel with the new during a transition
period. This is to avoid problems with API breakages. During this
period, ReportedDelay is disabled or DelayAgnostic is enabled, DA-AEC
is engaged in APM.
BUG=webrtc:4651
R=bjornv@webrtc.org, tommi@webrtc.org
Review URL: https://codereview.webrtc.org/1211053006
Cr-Commit-Position: refs/heads/master@{#9531}
Original review at https://codereview.webrtc.org/1180423006
SystemDelayTests was not updated w.r.t. extended_filter mode and some tests were disabled on Android since DA-AEC is automatically set.
All tests have now been updated for both extended_filter mode as well as DA-AEC, hence are now enabled on Android.
Also
* Moves default settings of extended_filter and DA-AEC form Init() to Create() to avoid unintentional loss of state during a reset.
* Fixes a potential bug of starting from scratch in extended_filter mode + DA-AEC.
This reverts commit 01c9b012e9171c813ace9e405c32fc75f4262bf6.
BUG=
R=henrik.lundin@webrtc.org
Review URL: https://codereview.webrtc.org/1187943005.
Cr-Commit-Position: refs/heads/master@{#9458}
The code only affects DA-AEC, but since DA-AEC is the default AEC if run on Android tests failed. Reverting to fix that test.
This reverts commit 9002cc426dab7a576f5247f45ba888cd081a39f0.
BUG=
TBR=henrik.lundin@webrtc.org
Review URL: https://codereview.webrtc.org/1183243003.
Cr-Commit-Position: refs/heads/master@{#9453}
We've seen that if we get a buffer underrun followed by a sudden buffer build up the DA-AEC can't really catch up even though it should be possible to estimate the upcoming difference. We have a feature for this already, but that is only used in the regular AEC. This CL turns that feature on also for DA-AEC.
- Adds a helper function MoveFarReadPtrWithoutSystemDelayUpdate()
- Only apply conservative correction for positive delays, where we can put the AEC into a non-causal state
- Stuff the farend buffer if we don't have enough data to process w.r.t. to current nearend buffer.
- Always run delay estimation based on reported delays to catch buffer starvation.
BUG=
R=henrik.lundin@webrtc.org
Review URL: https://codereview.webrtc.org/1180423006.
Cr-Commit-Position: refs/heads/master@{#9452}
This includes changes like:
* Attempt to break lines at better positions
* Use "override" in more places, don't use "virtual" with it
* Use {} where the body is more than one line
* Make declaration and definition arg names match
* Eliminate unused code
* EXPECT_EQ(expected, actual) (but use (actual, expected) for e.g. _GT)
* Correct #include order
* Use anonymous namespaces in preference to "static" for file-scoping
* Eliminate unnecessary casts
* Update reference code in comments of ARM assembly sources to match actual current C code
* Fix indenting to be more style-guide compliant
* Use arraysize() in more places
* Use bool instead of int for "boolean" values (0/1)
* Shorten and simplify code
* Spaces around operators
* 80 column limit
* Use const more consistently
* Space goes after '*' in type name, not before
* Remove unnecessary return values
* Use "(var == const)", not "(const == var)"
* Spelling
* Prefer true, typed constants to "enum hack" constants
* Avoid "virtual" on non-overridden functions
* ASSERT(x == y) -> ASSERT_EQ(y, x)
BUG=none
R=andrew@webrtc.org, asapersson@webrtc.org, henrika@webrtc.org, juberti@webrtc.org, kjellander@webrtc.org, kwiberg@webrtc.org
Review URL: https://codereview.webrtc.org/1172163004
Cr-Commit-Position: refs/heads/master@{#9420}
(This reverts commit 3fbf3f8841b5460503fb646eaedcb063620434a8.)
The original submission was reverted because it broke the Chrome build. This is fixed in patch set 2 of this change by keeping the old MediaConstraintsInterface string kExperimentalEchoCancellation. It will be removed once the Chrome code has been updated.
Original description:
"We use this Config struct for enabling/disabling Extended filter mode in AEC. This change renames it to ExtendedFilter for readability reasons. The corresponding media constraint is also renamed to kExtendedFilterEchoCancellation.
The old Config is kept in parallel with the new during a transition period. This is to avoid problems with API breakages. During this period, if any of the two Configs are enabled, the extended filter mode is engaged in APM. That is, the two Configs are combined with an "OR" operation.
This change also renames experimental_aec in AudioOptions to extended_filter_aec."
BUG=webrtc:4696
R=bjornv@webrtc.org, tommi@webrtc.org
Review URL: https://codereview.webrtc.org/1151573021.
Cr-Commit-Position: refs/heads/master@{#9401}
We use this Config struct for enabling/disabling Extended filter mode
in AEC. This change renames it to ExtendedFilter for readability
reasons. The corresponding media constraint is also renamed to
kExtendedFilterEchoCancellation.
The old Config is kept in parallel with the new during a transition
period. This is to avoid problems with API breakages. During this
period, if any of the two Configs are enabled, the extended filter
mode is engaged in APM. That is, the two Configs are combined with an
"OR" operation.
This change also renames experimental_aec in AudioOptions to extended_filter_aec.
BUG=4696
R=bjornv@webrtc.org, tommi@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/54659004
Cr-Commit-Position: refs/heads/master@{#9378}
When running AEC in extended_filter mode there is no startup phase to evaluate the reported system delay values.
Instead we simply use the first value and scale by two to avoid over compensating when synchronizing render and capture.
We don't need to be too accurate since we have extended the filter length.
On Android we use fixed (measured) reported delay values.
There is no need to be extra conservative here, because that is already built-in in the measured value.
In fact, the difference between devices is large and with such an extra conservative approach the true delay can not be caught by the filter length.
With this change we can improve performance on some devices.
BUG=4472
TESTED=offline on recordings from various devices
R=kwiberg@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/49909004
Cr-Commit-Position: refs/heads/master@{#9144}
There is no point in returning an error when Free() fails. In fact it can only happen if we have a null pointer as object. There is further no place where the return value is used.
Affected components are
- aec
- aecm
- agc
- ns
BUG=441
R=kwiberg@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/50579004
Cr-Commit-Position: refs/heads/master@{#8966}
To more easily determine if for example the AEC is not working properly one could monitor how often the estimated delay is out of bounds. With out of bounds we mean either being negative or too large, where both cases will break the AEC.
A new delay metric is added telling the user how often poor delay values were estimated. This is measured in percentage since last time the metrics were calculated.
All APIs have been updated with a third parameter with EchoCancellation::GetDelayMetrics() giving the option to exclude the new metric not to break existing code.
The new metric has been added to audio_processing_unittests with an additional protobuf member, and reference files accordingly updated.
voe_auto_test has not been updated to display the new metric.
BUG=4246
TESTED=audioproc on files
R=aluebs@webrtc.org, andrew@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/39739004
Cr-Commit-Position: refs/heads/master@{#8230}
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8230 4adac7df-926f-26a2-2b94-8c16560cd09d
Doing something similar for the band 16-24kHz to what is done for the band 8-16kHz. The only difference is that there is no comfort noise added in this band. Could not test how this sounds because there are no aecdumps with 48kHz sample rate as nfar as I know.
Tested for 32kHz sample rate and the output is bitexact with how it was before this CL.
BUG=webrtc:3146
R=andrew@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/28319004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@8080 4adac7df-926f-26a2-2b94-8c16560cd09d
audio_processing did not compile when aec_untrusted_delay_for_testing=1 was set. The constant kDelayDiffOffsetSamples was declared only for Mac when WEBRTC_UNTRUSTED_DELAY was automatically turned on.
Moving the declaration outside the ifdef makes it build with the flag on for any platform.
BUG=3673
TESTED=locally and trybots
R=andrew@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/22049004
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6866 4adac7df-926f-26a2-2b94-8c16560cd09d
Each audio processing step is given a pointer to an AudioBuffer, where
it can read and write int data. This patch adds corresponding
AudioBuffer methods to read and write float data; the buffer will
automatically convert the stored data between int and float as
necessary.
This patch also modifies the echo cancellation step to make use of the
new methods (it was already using floats internally; now it doesn't
have to convert from and to ints anymore).
(The reference data to the ApmTest.Process test had to be modified
slightly; this is because the echo canceller no longer unnecessarily
converts float data to int and then immediately back to float for each
iteration in the loop in EchoCancellationImpl::ProcessCaptureAudio.)
BUG=
R=aluebs@webrtc.org, andrew@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/18399005
git-svn-id: http://webrtc.googlecode.com/svn/trunk@6138 4adac7df-926f-26a2-2b94-8c16560cd09d
The untrusted delay mode provides an option to reduce the "known delay"
parameter passed down to the core AEC. This is necessary to handle the
very low latencies observed with the Chromium audio backend on Mac.
Prior to this change, it was possible to pass a negative value. The AEC
produced good output in practice, but it turned out this tripped a
heretofore unnoticed assert in ProcessBlock().
This change avoids the assert, and maintains the good output across a
set of Mac recordings. Bit-exact in some cases, and in the remaining,
quickly converging to identical output.
The assert was hit on the last webrtc roll in Chromium in
content_browsertests on Mac.
Corresponds to:
https://chromereviews.googleplex.com/9960013
TBR=bjornv
TESTED=Verified locally that "content_browsertests
--gtest_filter=WebrtcBrowserTest.*"" passes.
Review URL: https://webrtc-codereview.appspot.com/2328005
git-svn-id: http://webrtc.googlecode.com/svn/trunk@4886 4adac7df-926f-26a2-2b94-8c16560cd09d
Re-land: http://review.webrtc.org/2151007/TBR=bjornv@webrtc.org
Original change description:
This mode extends the filter length from the current 48 ms to 128 ms.
It is runtime selectable which allows it to be enabled through
experiment. We reuse the DelayCorrection infrastructure to avoid having
to replumb everything up to libjingle.
Increases AEC complexity by ~50% on modern x86 CPUs.
Measurements (in percent of usage on one core):
Machine/CPU Normal Extended
MacBook Retina (Early 2013),
Core i7 Ivy Bridge (2.7 GHz, hyperthreaded) 0.6% 0.9%
MacBook Air (Late 2010), Core 2 Duo (2.13 GHz) 1.4% 2.7%
Chromebook Pixel, Core i5 Ivy Bridge (1.8 GHz) 0.6% 1.0%
Samsung ARM Chromebook,
Samsung Exynos 5 Dual (1.7 GHz) 3.2% 5.6%
The relative value is large of course but the absolute should be
acceptable in order to have a working AEC on some platforms.
Detailed changes to the algorithm:
- The filter length is changed from 48 to 128 ms. This comes with tuning
of several parameters: i) filter adaptation stepsize and error
threshold; ii) non-linear processing smoothing and overdrive.
- Option to ignore the reported delays on platforms which we deem
sufficiently unreliable. Currently this will be enabled in Chromium for
Mac.
- Faster startup times by removing the excessive "startup phase"
processing of reported delays.
- Much more conservative adjustments to the far-end read pointer. We
smooth the delay difference more heavily, and back off from the
difference more. Adjustments force a readaptation of the filter, so they
should be avoided except when really necessary.
Corresponds to these changes:
https://chromereviews.googleplex.com/9412014https://chromereviews.googleplex.com/9514013https://chromereviews.googleplex.com/9960013
BUG=454,827,1261
Review URL: https://webrtc-codereview.appspot.com/2295006
git-svn-id: http://webrtc.googlecode.com/svn/trunk@4848 4adac7df-926f-26a2-2b94-8c16560cd09d
> Add an extended filter mode to AEC.
>
> This mode extends the filter length from the current 48 ms to 128 ms.
> It is runtime selectable which allows it to be enabled through
> experiment. We reuse the DelayCorrection infrastructure to avoid having
> to replumb everything up to libjingle.
>
> Increases AEC complexity by ~50% on modern x86 CPUs.
> Measurements (in percent of usage on one core):
>
> Machine/CPU Normal Extended
> MacBook Retina (Early 2013),
> Core i7 Ivy Bridge (2.7 GHz, hyperthreaded) 0.6% 0.9%
>
> MacBook Air (Late 2010), Core 2 Duo (2.13 GHz) 1.4% 2.7%
>
> Chromebook Pixel, Core i5 Ivy Bridge (1.8 GHz) 0.6% 1.0%
>
> Samsung ARM Chromebook,
> Samsung Exynos 5 Dual (1.7 GHz) 3.2% 5.6%
>
> The relative value is large of course but the absolute should be
> acceptable in order to have a working AEC on some platforms.
>
> Detailed changes to the algorithm:
> - The filter length is changed from 48 to 128 ms. This comes with tuning
> of several parameters: i) filter adaptation stepsize and error
> threshold; ii) non-linear processing smoothing and overdrive.
> - Option to ignore the reported delays on platforms which we deem
> sufficiently unreliable. Currently this will be enabled in Chromium for
> Mac.
> - Faster startup times by removing the excessive "startup phase"
> processing of reported delays.
> - Much more conservative adjustments to the far-end read pointer. We
> smooth the delay difference more heavily, and back off from the
> difference more. Adjustments force a readaptation of the filter, so they
> should be avoided except when really necessary.
>
> Corresponds to these changes:
> https://chromereviews.googleplex.com/9412014
> https://chromereviews.googleplex.com/9514013
> https://chromereviews.googleplex.com/9960013
>
> BUG=454,827,1261
> R=bjornv@webrtc.org
>
> Review URL: https://webrtc-codereview.appspot.com/2151007TBR=andrew@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/2296005
git-svn-id: http://webrtc.googlecode.com/svn/trunk@4839 4adac7df-926f-26a2-2b94-8c16560cd09d
This mode extends the filter length from the current 48 ms to 128 ms.
It is runtime selectable which allows it to be enabled through
experiment. We reuse the DelayCorrection infrastructure to avoid having
to replumb everything up to libjingle.
Increases AEC complexity by ~50% on modern x86 CPUs.
Measurements (in percent of usage on one core):
Machine/CPU Normal Extended
MacBook Retina (Early 2013),
Core i7 Ivy Bridge (2.7 GHz, hyperthreaded) 0.6% 0.9%
MacBook Air (Late 2010), Core 2 Duo (2.13 GHz) 1.4% 2.7%
Chromebook Pixel, Core i5 Ivy Bridge (1.8 GHz) 0.6% 1.0%
Samsung ARM Chromebook,
Samsung Exynos 5 Dual (1.7 GHz) 3.2% 5.6%
The relative value is large of course but the absolute should be
acceptable in order to have a working AEC on some platforms.
Detailed changes to the algorithm:
- The filter length is changed from 48 to 128 ms. This comes with tuning
of several parameters: i) filter adaptation stepsize and error
threshold; ii) non-linear processing smoothing and overdrive.
- Option to ignore the reported delays on platforms which we deem
sufficiently unreliable. Currently this will be enabled in Chromium for
Mac.
- Faster startup times by removing the excessive "startup phase"
processing of reported delays.
- Much more conservative adjustments to the far-end read pointer. We
smooth the delay difference more heavily, and back off from the
difference more. Adjustments force a readaptation of the filter, so they
should be avoided except when really necessary.
Corresponds to these changes:
https://chromereviews.googleplex.com/9412014https://chromereviews.googleplex.com/9514013https://chromereviews.googleplex.com/9960013
BUG=454,827,1261
R=bjornv@webrtc.org
Review URL: https://webrtc-codereview.appspot.com/2151007
git-svn-id: http://webrtc.googlecode.com/svn/trunk@4837 4adac7df-926f-26a2-2b94-8c16560cd09d
