| commit | 7bd5f253bc09fe18bc4190fe88dbba9792ce32ae | [log] [tgz] |
|---|---|---|
| author | aluebs <aluebs@webrtc.org> | Tue Jun 21 11:30:25 2016 -0700 |
| committer | Commit bot <commit-bot@chromium.org> | Tue Jun 21 18:30:31 2016 +0000 |
| tree | 4657421d8a2680336a2198a79b0131a196ed3e21 | |
| parent | 435264a183e16572c99e8dd46171163b6fd33d59 [diff] [blame] |
Fine tune the IntelligibilityEnhancer Label less chunks as speech, adapt slower and be more conservative with the maximum gain it can apply. Review-Url: https://codereview.webrtc.org/2087623003 Cr-Commit-Position: refs/heads/master@{#13242}
diff --git a/webrtc/modules/audio_processing/intelligibility/intelligibility_utils.cc b/webrtc/modules/audio_processing/intelligibility/intelligibility_utils.cc index 3675f66..dfa912b 100644 --- a/webrtc/modules/audio_processing/intelligibility/intelligibility_utils.cc +++ b/webrtc/modules/audio_processing/intelligibility/intelligibility_utils.cc
@@ -23,7 +23,7 @@ namespace { const float kMinFactor = 0.01f; -const float kMaxFactor = 1000.f; +const float kMaxFactor = 100.f; // Return |current| changed towards |target|, with the relative change being at // most |limit|.