- From: Raymond Toy <rtoy@google.com>
- Date: Thu, 7 Jan 2021 10:20:47 -0800
- To: "public-audio@w3.org Group" <public-audio@w3.org>, public-audio-comgp@w3.org
- Message-ID: <CAE3TgXGm9862AD9SC8p4FOBb88MdSdwZtvmpN273HH+Li8W+fA@mail.gmail.com>
Attendees: Hongchan Choi, Jeff Switzer, Athanasio Malamos, Paul Adenot,
Raymond Toy, Ruth John
Minutes:
- New Patent Policy
<https://www.w3.org/Consortium/Patent-Policy-20200915/> for WebAudio
- Royalty-free licensing for all group members
- Disclose essential claims if requested
- Ask a lawyer if you're not sure.
- V1
- CR Status
- Chris Lilley not here, so nothing to do.
- New Issues
- Request to Expose AudioBuffer to DedicatedWorker
<https://github.com/WebAudio/web-audio-api/issues/2288>
- Raymond: What do they really want?
- Paul, Hongchan: Just need a way to hold audio
- Paul: Say, you get an mp3 in 16-bit interleaved. That needs
to be converted to AudioBuffer float/planar, and then
converted back
eventually to convert back to mp3, say.
- Paul: No way to do that now; moving to v2.
- Hongchan: Good point
- Raymond: Still want to know the use case and whether
AudioBuffer is appropriate. Could there be something better suited?
- Paul: I'll also talk to Chris Cunningham in the next webcodec
meeting.
- Inconsistencies with "control thread state"
<https://github.com/WebAudio/web-audio-api/issues/2280>
- Raymond: Opened in Dec, but we didn't discuss it.
- Raymond: I think it's basically right and we just need to
review the uses and make it consistent.
- Paul: Yes.
- Outstanding issues
- ABSN.start control messages
<https://github.com/WebAudio/web-audio-api/issues/2248>
- Raymond: PR ready to clarify what this is trying to do: ABSN
can resume a paused context.
- Paul: I'll add myself as a reviewer
- Invoking JS in the rendering loop needs a bunch of setup that
doesn't seem to be in the spec
<https://github.com/WebAudio/web-audio-api/issues/1967> and
Integration
between event-loop of AudioWorkletGlobalScope and rendering loop
<https://github.com/WebAudio/web-audio-api/issues/2008>
- Paul: Still in progress. Explains when onmessage and process
gets called.
- Paul: Depending on the back end and computer, you can have
various results.
- PRs
- Fix #2248: Clean up ABSN start algorithm
<https://github.com/WebAudio/web-audio-api/pull/2286>
- Fix #2281: CR 2020-12-15
<https://github.com/WebAudio/web-audio-api/pull/2283>
- V2
- New issues
- Should 'Atomics.wait' be available in AudioWorklets associated
with an OfflineAudioContext?
<https://github.com/WebAudio/web-audio-api-v2/issues/110>
- Paul: Nice to keep them the same.
- Raymond: I agree
- Raymond: Could suspend/resume work?
- Hongchan: Probably doesn't work for their use case.
- Raymond: Conclusion: Won't do it?
- Paul: Yes.
- [Paul to update issue]
- Alternatives for module loading of AudioWorklet
<https://github.com/WebAudio/web-audio-api-v2/issues/109>
- Paul: This is really a worklet problem and more specifically a
script loading problem.
- Paul: Github sets CSP so you can't run things in the console.
- Hongchan: What is hte alternative here? Blobs?
- Paul: I think I do this a lot
- Hongchan: This is CSP specific. It should work if he runs
this on his own server.
- [Paul to comment]
- Hongchan: We're not the right people; it should be handled by
Worklet.
- Issues
- NoiseGenerator
<https://github.com/WebAudio/web-audio-api-v2/issues/8>: Updates
on using bigint
- Paul: No updates
- priority-1 issues
<https://github.com/WebAudio/web-audio-api-v2/issues?q=is%3Aopen+is%3Aissue+label%3Apriority-1>
- Select audio device
<https://github.com/WebAudio/web-audio-api-v2/issues/10>, get
device change notification, audio config
- Raymond: We ended the last meeting with what happens when
devices are plugged or unplugged
- Raymond: Seems application-specific
- Hongchan: Do we want to expose and implement this?
- Paul: We could provide events on things changing so
developers can decide what to do.
- [Raymond to update issue]
- Headphone detection
<https://github.com/WebAudio/web-audio-api-v2/issues/89>
- Hongchan: See issue for more details.
https://github.com/WebAudio/web-audio-api-v2/issues/89#issuecomment-756234076
- Raymond: We used to have other things we wanted. Can we add
that now.
- Hongchan: Perhaps it was audio output?
- Raymond: Maybe; don't remember
- API proposal to preemptively determine audio thread overload (to
avoid audio glitching)
<https://github.com/WebAudio/web-audio-api-v2/issues/40>
- Paul: Using Firefox's tools, sometimes I can see the CPU is ok,
but peaks every 3rd render.
- Raymond: That's a problem we we return the usage of each
render. The user could just miss all the bad spots, so
you think ithere's
still CPU but sounds crappy.
- Paul: That happens quite frequently.
- Honghchan: I was thinking of returning what chromes devtool
does
- Raymond: But that's already smoothed.
- Raymond: What's really important is the peak
- Paul: Yes, like a vumeter
- Honghchan: Right. We'll need some way of resetting the peak
- Hongchan: Is 4 buckets enough?
- Raymond: I think that's too coarse. There's a big difference
between 3/4 and 1. Perhaps 10 buckets? If you're at 90%
that's close
enough to 100% that you need to be careful
- [Times up]
- [Honghan to update issue]
See y'all next week!
On Wed, Jan 6, 2021 at 3:35 PM Raymond Toy <rtoy@google.com> wrote:
> Here's the agenda for our WG/CG meeting, the first of the new year!
>
> Meeting venue https://meet.google.com/wgg-wjtd-kdu
>
> Agenda:
>
> - New Patent Policy
> <https://www.w3.org/Consortium/Patent-Policy-20200915/> for WebAudio
> - Royalty-free licensing for all group members
> - Disclose essential claims if requested
> - Ask a lawyer if you're not sure.
> - V1
> - CR Status
> - New Issues
> - Request to Expose AudioBuffer to DedicatedWorker
> <https://github.com/WebAudio/web-audio-api/issues/2288>
> - Inconsistencies with "control thread state"
> <https://github.com/WebAudio/web-audio-api/issues/2280>
> - Outstanding issues
> - ABSN.start control messages
> <https://github.com/WebAudio/web-audio-api/issues/2248>
> - Invoking JS in the rendering loop needs a bunch of setup that
> doesn't seem to be in the spec
> <https://github.com/WebAudio/web-audio-api/issues/1967> and Integration
> between event-loop of AudioWorkletGlobalScope and rendering loop
> <https://github.com/WebAudio/web-audio-api/issues/2008>
> - PRs
> - Fix #2248: Clean up ABSN start algorithm
> <https://github.com/WebAudio/web-audio-api/pull/2286>
> - Fix #2281: CR 2020-12-15
> <https://github.com/WebAudio/web-audio-api/pull/2283>
> - V2
> - New issues
> - Should 'Atomics.wait' be available in AudioWorklets associated
> with an OfflineAudioContext?
> <https://github.com/WebAudio/web-audio-api-v2/issues/110>
> - Alternatives for module loading of AudioWorklet
> <https://github.com/WebAudio/web-audio-api-v2/issues/109>
> - Issues
> - NoiseGenerator
> <https://github.com/WebAudio/web-audio-api-v2/issues/8>: Updates
> on using bigint
> - priority-1 issues
> <https://github.com/WebAudio/web-audio-api-v2/issues?q=is%3Aopen+is%3Aissue+label%3Apriority-1>
> - Select audio device
> <https://github.com/WebAudio/web-audio-api-v2/issues/10>, get
> device change notification, audio config
> - Others?
> - priority-2 issues
> <https://github.com/WebAudio/web-audio-api-v2/issues?q=is%3Aopen+is%3Aissue+sort%3Aupdated-desc+label%3Apriority-2>
> - PRs
> - Bikeshed boilerplate
> <https://github.com/WebAudio/web-audio-api-v2/pull/95>
> - AOB
>
>
Received on Thursday, 7 January 2021 18:21:13 UTC