diff --git a/CHANGELOG.md b/CHANGELOG.md
index 2f780b206..3d302dffc 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -26,8 +26,9 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 ### Changed
 - HFP: API functions return status code if appropriate 
 - HFP: removed one parameter from hfp_ag/hf_activate_voice_recognition function, instead introduced hfp_ag/hf_deactivate_voice_recognition 
-- HFP: event HFP_SUBEVENT_VOICE_RECOGNITION_STATUS emitted on (enhanced) voice recognition status change. Field "activated" replaced with "state". Added additional "status" field that indicates if the command was successful.
-- HFP: streamlined enhanced voice recognition function names
+- HFP: merged legacy and enhanced Voice Recognition Activation (VRA) API. If available enhanced mode will be preferred.
+- HFP: seperate events for acivation (HFP_SUBEVENT_VOICE_RECOGNITION_ACTIVATED) and deactivation (HFP_SUBEVENT_VOICE_RECOGNITION_DEACTIVATED) instead of combined HFP_SUBEVENT_VOICE_RECOGNITION_STATUS with status field. HFP_SUBEVENT_VOICE_RECOGNITION_ACTIVATED contains field "enhanced" to indicate if enhanced mode is used. 
+- HFP, enhanced VRA: HFP_SUBEVENT_VOICE_RECOGNITION_ACTIVATED is emitted after VRA is ready and the the audio connection is established. This simplifies HFP HF client logic, i.e. client can call `hfp_hf_enhanced_voice_recognition_report_ready_for_audio directly` upon reception of HFP_SUBEVENT_VOICE_RECOGNITION_ACTIVATED event.
 - AVDTP: media config validator is called with preview of media codec configuration event and configured separately for sink/source
 - AVRCP: use PANEL as default unit + subunit info
 - Run Loop: new functionality for HCI transport drivers and inter-process communication