Add audioLevel to RTCEncodedAudioFrameMetadata

[guidou]

This allows applications that use encoded transform to get the audioLevel directly from the encoded frames instead of having to poll getContributingSources or getStats. This would reduce resource consumption and would be more accurate.

Some have also mentioned using the audio level to control audio redundancy, by reducing redundancy for (i.e., dropping) frames with zero/low audio level.

[dontcallmedom-bot]

This issue had an associated resolution in WebRTC April 2025 meeting – 22 April 2025 (Issue #244 - Add audioLevel to RTCEncodedAudioFrameMetadata):

RESOLUTION: prepare a PR