This CL changes the name of classes, methods and variables making using "noise generator".
This naming is replaced with "input-reference generator" which is more descriptive of the actual role.
Comments, CSS class and HTML item names have also been changed.
Consistency for variable names has been verified and the style checked with pylint.
BUG=webrtc:7218
Review-Url: https://codereview.webrtc.org/2805653002
Cr-Commit-Position: refs/heads/master@{#17639}
Changes in the microphone gain are effecting the AEC in the sense
that each change in the microphone gain is a change in the echo
path seen by the AEC. This CL utilizes the ability of AEC3 to
leverage information about known changes in the analog microphone
gain.
BUG=webrtc:6018
Review-Url: https://codereview.webrtc.org/2808073002
Cr-Commit-Position: refs/heads/master@{#17625}
During the first few capture frames, there is no way for the AEC
to tell whether there is echo in the capture signal as the echo
removal functionality in the AEC has not yet seen any render
signal. To avoid initial echo bursts due to this, this CL adds
functionality for forcing the echo suppression gain to zero during
the first 50 blocks (200 ms) after call start and after a reported
echo path change.
BUG=webrtc:6018
Review-Url: https://codereview.webrtc.org/2808733002
Cr-Commit-Position: refs/heads/master@{#17624}
The unit test ConversationalSpeechTest.MultiEndCallWavReaderAdaptorSine uses CreateSineWavFile() and writes temporary wav files that are used for test (deleted only if the test passes).
BUG=webrtc:7218
Review-Url: https://codereview.webrtc.org/2774423005
Cr-Commit-Position: refs/heads/master@{#17608}
This CL includes extensive tests to match accept or reject decisions on several different timing setups. The setups are simulated using mocks (by far more light-weight than using actual timing and audio track files).
The client code, the unit tests in this case, passes information about the fake audio tracks to MockWavReaderFactory. MockWavReader instances are then created using the parameters defined in the client code. To improve the readability of the tests, generator_unittest.cc includes a docstring explaining how each MultiEndCallSetup* test is documented.
Run tests as follows:
$ out/Default/modules_unittests --gtest_filter=ConversationalSpeechTest.*
BUG=webrtc:7218
Review-Url: https://codereview.webrtc.org/2781573002
Cr-Commit-Position: refs/heads/master@{#17592}
This CL adds support for handling highly reverberant echoes in
AEC3. The functionality is hardcoded to be have no effect (via
a decay factor of 0), but this CL will be followed by other CLs
for which nonzero decay factors will be allowed.
Apart from this change, this CL also refactors the residual
echo estimation code to make it shorter and more readable.
The changes introduced herein are bitexact (for a decay factor
of 0).
BUG=webrtc:6018
Review-Url: https://codereview.webrtc.org/2804223002
Cr-Commit-Position: refs/heads/master@{#17589}
This CL ensures that the number of bands
for the render side matches that for the capture side
when AEC3 is active. Without this, there was problems
when the render rate is different from the capture rate.
BUG=webrtc:6018
Review-Url: https://codereview.webrtc.org/2800033003
Cr-Commit-Position: refs/heads/master@{#17586}
Reason for revert:
Trying to re-land after solving some related issues.
There are no changes compared to the original CL.
Original issue's description:
> Revert of Loosening the coupling between WebRTC and //third_party/protobuf (patchset #16 id:300001 of https://codereview.webrtc.org/2747863003/ )
>
> Reason for revert:
> I will try to reland next week because it is causing some problems.
>
> Original issue's description:
> > To accommodate some downstream WebRTC users we need to loosen
> > the coupling between our code and the //third_party/protobuf.
> >
> > This includes using typedefs to define strings instead of
> > assuming std::string.
> >
> > After this refactoring it will be possible to link with other
> > protobuf implementations than the current one.
> >
> > We moved the PRESUBMIT check to another CL [1]. The goal of this
> > presubmit is to avoid the direct usage of google::protobuf outside
> > of the webrtc/base/protobuf_utils.h header file.
> >
> > [1] - https://codereview.webrtc.org/2753823003/
> >
> > BUG=webrtc:7340
> > NOTRY=True
> >
> > Review-Url: https://codereview.webrtc.org/2747863003
> > Cr-Commit-Position: refs/heads/master@{#17466}
> > Committed: 16ab93b952
>
> TBR=kjellander@webrtc.org,henrik.lundin@webrtc.org,kwiberg@webrtc.org,michaelt@webrtc.org,peah@webrtc.org
> # Not skipping CQ checks because original CL landed more than 1 days ago.
> BUG=webrtc:7340
>
> Review-Url: https://codereview.webrtc.org/2786363002
> Cr-Commit-Position: refs/heads/master@{#17483}
> Committed: d00aad5eb2TBR=kjellander@webrtc.org,henrik.lundin@webrtc.org,kwiberg@webrtc.org,michaelt@webrtc.org,peah@webrtc.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=webrtc:7340
NOTRY=True
Review-Url: https://codereview.webrtc.org/2791963003
Cr-Commit-Position: refs/heads/master@{#17584}
This CL corrects the behavior during buffer underruns.
Furthermore, it increases the tolerance for API call jitter, and
removes the minimum value for the comfort noise.
BUG=webrtc:6018
Review-Url: https://codereview.webrtc.org/2787123003
Cr-Commit-Position: refs/heads/master@{#17576}
This CL adds fairly significant changes to the echo removal
functionality, the main ones being.
-More centralized control over the echo removal.
-Updated echo suppression gain behavior.
-Significantly increased usage of the linear adaptive filter.
-New echo removal functionality when the linear filter is not usable.
This CL is chained to the CL https://codereview.webrtc.org/2784023002/
BUG=webrtc:6018
Review-Url: https://codereview.webrtc.org/2782423003
Cr-Commit-Position: refs/heads/master@{#17575}
This CL adds major render pipeline changes to the AEC3 code. The reason
for these are that
1) It allows the echo removal unit to receive information about the content
in bands beyond band 0, thereby allowing removal of high-frequency
echoes
2) It allows more controlled handling of the render buffers, allowing proper
buffer behaviour during capture glitches and clock-drift.
Unfortunately, the render pipeline caused a lot of related changes in much
of the rest of the AEC3 files. Most of these are, however, caused by
a change of class name.
Another unfortunate effect of this CL, is that a number of unittest cease to
compile. I chose to temporarily solve that by removing them from the
build using #if/#endif. The reason for that is that those will anyway again
need to be changed in the next review, and doing like this avoids them
having to be reviewed twice.
BUG=webrtc:6018
Review-Url: https://codereview.webrtc.org/2784023002
Cr-Commit-Position: refs/heads/master@{#17547}
- snake_case -> CapWords
- compulsory docstring added
- style
A followup CL will fix remaining issues as raised by the next version of the WebRTC Python linter (update in progress).
BUG=webrtc:7218
NOTRY=True
Review-Url: https://codereview.webrtc.org/2793903006
Cr-Commit-Position: refs/heads/master@{#17543}
Reason for revert:
Issue with rebasing from depending CL
Original issue's description:
> Export library that generates an HTLM file with the scores organized in tables.
>
> BUG=webrtc:7218
> NOTRY=True
>
> Review-Url: https://codereview.webrtc.org/2717973006
> Cr-Commit-Position: refs/heads/master@{#17512}
> Committed: 29e3330139TBR=aleloi@webrtc.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=webrtc:7218
Review-Url: https://codereview.webrtc.org/2791293002
Cr-Commit-Position: refs/heads/master@{#17513}
This CL also includes the implementation of the echo noise generator.
I added comments in the diff to help the reviewers.
BUG=webrtc:7218
NOTRY=True
Review-Url: https://codereview.webrtc.org/2715233003
Cr-Commit-Position: refs/heads/master@{#17507}
Reason for revert:
I will try to reland next week because it is causing some problems.
Original issue's description:
> To accommodate some downstream WebRTC users we need to loosen
> the coupling between our code and the //third_party/protobuf.
>
> This includes using typedefs to define strings instead of
> assuming std::string.
>
> After this refactoring it will be possible to link with other
> protobuf implementations than the current one.
>
> We moved the PRESUBMIT check to another CL [1]. The goal of this
> presubmit is to avoid the direct usage of google::protobuf outside
> of the webrtc/base/protobuf_utils.h header file.
>
> [1] - https://codereview.webrtc.org/2753823003/
>
> BUG=webrtc:7340
> NOTRY=True
>
> Review-Url: https://codereview.webrtc.org/2747863003
> Cr-Commit-Position: refs/heads/master@{#17466}
> Committed: 16ab93b952TBR=kjellander@webrtc.org,henrik.lundin@webrtc.org,kwiberg@webrtc.org,michaelt@webrtc.org,peah@webrtc.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=webrtc:7340
Review-Url: https://codereview.webrtc.org/2786363002
Cr-Commit-Position: refs/heads/master@{#17483}
the coupling between our code and the //third_party/protobuf.
This includes using typedefs to define strings instead of
assuming std::string.
After this refactoring it will be possible to link with other
protobuf implementations than the current one.
We moved the PRESUBMIT check to another CL [1]. The goal of this
presubmit is to avoid the direct usage of google::protobuf outside
of the webrtc/base/protobuf_utils.h header file.
[1] - https://codereview.webrtc.org/2753823003/
BUG=webrtc:7340
NOTRY=True
Review-Url: https://codereview.webrtc.org/2747863003
Cr-Commit-Position: refs/heads/master@{#17466}
Deletes left-over includes of trace.h and critical_section_wrapper.h.
BUG=webrtc:7035
Review-Url: https://codereview.webrtc.org/2784873002
Cr-Commit-Position: refs/heads/master@{#17460}
The command
tools/checklicenses/checklicenses.py --ignore-suppressions ./webrtc
previously produced this output:
'webrtc/modules/audio_processing/test/apmtest.m' has non-whitelisted license 'UNKNOWN'
'webrtc/modules/audio_processing/transient/test/readDetection.m' has non-whitelisted license 'UNKNOWN'
'webrtc/modules/audio_processing/transient/test/readPCM.m' has non-whitelisted license 'UNKNOWN'
...
This CL adds the WebRTC licence with appropriate year to all our
MATLAB files. All these files were contributed by WebRTC project
members hlundin@, pbos@, niklase@.
BUG=chromium:98592
NOTRY=True
Review-Url: https://codereview.webrtc.org/2781663005
Cr-Commit-Position: refs/heads/master@{#17428}
The class creates one WavReaderInterface object for each unique audiotrack and builds the set of speaker names.
Validating if the audiotrack lengths and the timing information are compatible (and hence valid) is not implemented yet.
MultiEndCall is designed using dependency injection. This allows to use mock objects with which we can quickly simulate different timings and track lengths without needing actual wav files.
BUG=webrtc:7218
Review-Url: https://codereview.webrtc.org/2761853002
Cr-Commit-Position: refs/heads/master@{#17421}
The ConversationalSpeechTest.* unit tests are now part of modules_unittests.
The rtc_test target has been replaced with an rtc_source_set one.
The latter is included as dependency in audio_processing_unittests.
BUG=webrtc:7218
Review-Url: https://codereview.webrtc.org/2769863005
Cr-Commit-Position: refs/heads/master@{#17360}
The conversational_speech::Timing class models a list of turns.
Each turn, is identified by a speaker, the audiotrack name, and an offset in milliseconds.
The unit test checks that an instance of Timing is correctly populated and that save/reload leads to the same data.
BUG=webrtc:7218
Review-Url: https://codereview.webrtc.org/2750353002
Cr-Commit-Position: refs/heads/master@{#17346}
The results of the echo detector complexity tests are currently notoriously spiky and unreliable. The following improvements are made in this CL:
- Significantly longer warmup time before starting the test
- More iterations and larger batches
- Different number of iterations for slow and fast tests
- Use the echo likelihood in the test so it cannot get optimized out
BUG=webrtc:7353
Review-Url: https://codereview.webrtc.org/2750413002
Cr-Commit-Position: refs/heads/master@{#17303}
Test for the conversational_speech::Config class and renaming.
BUG=webrtc:7218
Review-Url: https://codereview.webrtc.org/2749573002
Cr-Commit-Position: refs/heads/master@{#17301}
generation."
This CL removes the Python script and adds its C++ porting.
The former was in its early stage and it has permanently been removed.
This is a reland of https://codereview.webrtc.org/2740063004/ which
was reverted. Now the build errors are fixed.
BUG=webrtc:7218
Review-Url: https://codereview.webrtc.org/2752793002
Cr-Commit-Position: refs/heads/master@{#17300}
Currently no lock is taken when returning echo likelihood stats, which causes a race condition between the thread getting the stats and the thread running the echo detector. This CL resolves the issue by adding locking.
BUG=webrtc:7346
Review-Url: https://codereview.webrtc.org/2749973003
Cr-Commit-Position: refs/heads/master@{#17270}
Reason for revert:
Even if the conversational speech tool is external and not a core part of webrtc, there are too many trybots failing.
Original issue's description:
> C++ porting of the initial python script for conversational speech generation.
>
> This CL removes the Python script and adds its C++ porting.
> The former was in its early stage and it has permanently been removed.
>
> BUG=webrtc:7218
> NOTRY=True
>
> Review-Url: https://codereview.webrtc.org/2740063004
> Cr-Commit-Position: refs/heads/master@{#17254}
> Committed: 0cf3aa6d0dTBR=henrik.lundin@webrtc.org,kjellander@webrtc.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=webrtc:7218
Review-Url: https://codereview.webrtc.org/2753843002
Cr-Commit-Position: refs/heads/master@{#17257}
This CL removes the Python script and adds its C++ porting.
The former was in its early stage and it has permanently been removed.
BUG=webrtc:7218
NOTRY=True
Review-Url: https://codereview.webrtc.org/2740063004
Cr-Commit-Position: refs/heads/master@{#17254}
The only non-const operation was a one-time initialization of a member only used in this function. I've moved it to the ctor.
BUG=webrtc:5298
Review-Url: https://codereview.webrtc.org/2741733002
Cr-Commit-Position: refs/heads/master@{#17223}
It generates json files to be used as APM configuration presets.
Useful to avoid manual editing of such files.
BUG=webrtc:7218
NOTRY=True
Review-Url: https://codereview.webrtc.org/2719853006
Cr-Commit-Position: refs/heads/master@{#17013}
