[webrtc-pc] Don't require RTCRtpContributingSource.audioLevel to be present in some cases (#2046)

henbos has just created a new issue for https://github.com/w3c/webrtc-pc:

== Don't require RTCRtpContributingSource.audioLevel to be present in some cases ==
The [audioLevel](https://w3c.github.io/webrtc-pc/#dom-rtcrtpcontributingsource-audiolevel) is optional for CSRCs, but for SSRCs the spec says it MUST be present:

> For SSRCs, this MUST be converted from the level value defined in [RFC6464] if the RFC 6464 header extension is present, otherwise the user agent must compute the value from the audio data (the member must never be absent for audio receivers).

I think this implies that if you receive a packet, even if it arrives late and you intent to drop it, you have to waste cycles decoding it for the sake of calculating the audio level. I also wonder if there are performance considerations with having to calculate it.

Perhaps there should be some leeway about not always having to calculate this if the header extension is present? Chrome currently only surfaces this value from the header extension, not from calculating it.

Please view or discuss this issue at https://github.com/w3c/webrtc-pc/issues/2046 using your GitHub account

Received on Thursday, 10 January 2019 10:20:54 UTC