<hongchan> Okay - we're going to reconnect

<rtoyg_m> We're going to dial in again.

Finalization of AudioWorklet

joe - where do we stand?

hongchan I think we are pretty much done

prs ofr remaining issues are coming soon

<hongchan> https://github.com/WebAudio/web-audio-api/pull/1078

<hongchan> https://github.com/WebAudio/web-audio-api/pull/1078/commits/bf20a9d464eec69eec6f2ee95f86abc9ff8747d6

<hongchan> https://github.com/WebAudio/web-audio-api/pull/1078/commits/bf20a9d464eec69eec6f2ee95f86abc9ff8747d6

joe, is this already merged, hongchan slight difference between joes proposal and last pr

<hongchan> inputs[0].length > 0

<hongchan> inputs[0][0].length > 0

joe think this is correct

hongchan no more issues on audio worklet today

joe should say what happens if the process throws an exception

hongchan if process method throws, set active source to false. So no longer call processing

rtoyg_m this is not enough as if input is connected it active source would be ignored

joe just because an exception is thrown in 1 quantum does not mean it will continue to fail

hongchan one thing to consider is the animaton worklet approach, if pain or animate methods throw they are invalidated and no longer called

rtoyg_m throwing an exception on very render quantum is not ideal

<joe> https://github.com/WebAudio/web-audio-api/issues/1084

more thought required, issue raised ^^ discuss on list

#95 Paul's proposed spec for playbackRate interpretation

padenot bug in subsample accurate start and loop to be fixed, and a couple of edge cases.

joe found it difficult to tell which samplerate was being referenced

padenot runs at Acontext sr

padenot one concern, if you resample the buffer and convert the time, make sure there are no errors in the conversion.

joe offset, duration, loopstart/end in units of original buffer.

joe subsample interpretation, if there is a run of samples which are all fractionaly shifted.

padenot if starting from frational sample loop points must also be fractional

joe if a buffer is at the ac rate and does not need resampling playback rate of 1. loop start = 128 samples loopend 256.5 samples. duration is fractional. 1st time copying without interp after loop 0.5 frame off for each sample. Entire pass has to be interpolated to avoid glitching.

joe will loop alternate between being aligned and interpolated - should we keep interpolating until aligned again

padenot consider looping being a concat of the buffer - does not work at start and end.

joe usecase for fractional loops is align zero crossing in periodic audio. If we don't maintain the interp then we will not get the desired behaviour.

padenot will fix issues that have already been commented on and then improve for accurate looping

Determining path forward with respect to CR: what showstopper issues remain?

joe outstanding issues - specing the dynamics compression

rtoyg_m will work on it

<rtoyg_m> https://github.com/WebAudio/web-audio-api/issues/13

<rtoyg_m> side chain: https://github.com/WebAudio/web-audio-api/issues/246

mdjp we can use the cr process to determine if this is a must have for v1 without delaying further

<joe> https://github.com/WebAudio/web-audio-api/issues?q=is%3Aopen+is%3Aissue+milestone%3A%22Web+Audio+V1%22+label%3A%22Needs+WG+review%22

review v1 issues which need working group review

https://github.com/WebAudio/web-audio-api/issues/988

hongchan this can be closed

https://github.com/WebAudio/web-audio-api/issues/906

joe part of defining dynamics compressor?

hongchan I think is ideal to support native multichannel compression, but its too late for that.

padenot we can allow settable channel count, energy estimation is same for all channels.

padenot suggestion ^^

allow implementers to extend support if needed.

rtoyg_m propose to put this issue on hold until dynamics compressor is speced

<joe> https://github.com/WebAudio/web-audio-api/issues?utf8=%E2%9C%93&q=is%3Aopen%20is%3Aissue%20milestone%3A%22Web%20Audio%20V1%22%20label%3A%22Ready%20for%20Editing%22%20

Items for V1 that are ready for editing

<joe> https://github.com/WebAudio/web-audio-api/issues/264

Next meeting 1st December

trackbot, end meeting

This is scribe.perl Revision: 1.148  of Date: 2016/10/11 12:55:14  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

No ScribeNick specified.  Guessing ScribeNick: mdjp
Inferring Scribes: mdjp
Default Present: (no_one)
Present: (no_one)

WARNING: Fewer than 3 people found for Present list!


WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 17 Nov 2016
Guessing minutes URL: http://www.w3.org/2016/11/17-audio-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.