Re: echoCancelation (sic)

On 04/03/2014 07:16 AM, Randell Jesup wrote:
> On 4/2/2014 11:21 AM, Harald Alvestrand wrote:
>> On 04/02/2014 04:28 PM, Cullen Jennings (fluffy) wrote:
>>> three things
>>>
>>> 1) if it’s spelled wrong, lets fix it now - I don’t want to accept
>>> two spellings that is lame
>> agreed!
>>> 2) AEC needs tone on by default
>> "to be on", I guess
>>> 3) I seem to recall that the IETF specs required all WebRTC
>>> implementations to support AEC so not worried about case where
>>> something does not implement it.
>> Yep. From the use-cases-and-requirements draft, browser requirements:
>>
>>     F7     When there are both incoming and outgoing audio
>>             streams, echo cancellation must be made
>>             available to avoid disturbing echo during
>>             conversation.
>
> I'll note that's from rtcweb IIRC, which is not the same as
> getUserMedia() - but I agree the default should be AEC on to avoid
> surprises.  (I'm also fine if the UA knows there's no output stream
> available for AEC and disables AEC, there's no practical negative
> impact.)
>
It's from draft-ietf-rtcweb-use-cases-and-requirements, which was taken
as a shared IETF/W3C requirements statement, and applicable in W3C to
both getusermedia and to peerconnection. There's lots of other stuff in
there that has informed our discussion about getusermedia.

But we agree about the conclusion, so we don't have to debate the
premises much :-)



-- 
Surveillance is pervasive. Go Dark.

Received on Thursday, 3 April 2014 07:10:42 UTC