webrtc/modules/audio_processing/include
Per Åhgren 1f33a37565 AEC3 tunings to increase the transparency
This CL changes the tuning of AEC3 to increase the transparency.
In particular:
-The present parameters are re-tuned.
-An oversuppression factor is added in the newly added soft-knee in
 the NLP gain. The purpose of this is to avoid fluctuations in the
 residual echo.
-The dynamics of the computed gain are bounded to ensure that the 
 specified gain characteristics are realizable without echo leakage.
 This also adds robustness against echo leakage in frequency regions
 that are poorly estimated.
 This change was needed to avoid echo leakage from the above 
 tunings.

Bug: chromium:773543,webrtc:8378
Change-Id: If8acc41c1423a6a2fa6f8c4daf2735c86f0b529a
Reviewed-on: https://webrtc-review.googlesource.com/8262
Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
Commit-Queue: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20231}
2017-10-11 07:28:09 +00:00
..
aec_dump.cc Fixing WebRTC after moving from src/webrtc to src/ 2017-09-15 05:02:56 +00:00
aec_dump.h Fixing WebRTC after moving from src/webrtc to src/ 2017-09-15 05:02:56 +00:00
audio_processing.cc Fixing WebRTC after moving from src/webrtc to src/ 2017-09-15 05:02:56 +00:00
audio_processing.h AEC3 tunings to increase the transparency 2017-10-11 07:28:09 +00:00
config.cc Fixing WebRTC after moving from src/webrtc to src/ 2017-09-15 05:02:56 +00:00
config.h Fixing WebRTC after moving from src/webrtc to src/ 2017-09-15 05:02:56 +00:00
mock_audio_processing.h Add PostProcessing interface to audio processing module. 2017-09-26 14:07:15 +00:00