- From: Raymond Toy <rtoy@google.com>
- Date: Thu, 10 Dec 2020 10:08:08 -0800
- To: "public-audio@w3.org Group" <public-audio@w3.org>, public-audio-comgp@w3.org
- Message-ID: <CAE3TgXH0+Fcdr6HQaEQMD_+XkmvKYPGmj8w-f_EN2MM=HQ6rUg@mail.gmail.com>
Attendees: Hongchan Choi, Philippe Milot, Jeff Switzer, Raymond Toy, Paul
Adenot, Chris Lilley
Minutes:
- V1
- CR Status
- Raymond: Updated CR started; needs help from Chris Lilley
- Outstanding issues
- ABSN.start control messages
<https://github.com/WebAudio/web-audio-api/issues/2248>
- Raymond: No progress
- 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: Trying to sort things; need to do it carefully. Nothing
too crazy.
- PRs
- Fix #2032: ABSN extrapolates the last output
<https://github.com/WebAudio/web-audio-api/pull/2256>
- Raymond: Needs quick review from Paul
- V2
- New issues
- AudioOutputContext
<https://github.com/WebAudio/web-audio-api-v2/issues/106>
- Paul: Requires OS support. See getDisplayMedia. Can only be
implemented in Win, Linux and sometimes on Android. Chrome
has shipped it
- Honghcan: Would it be better in screen capture and not
WebAudio
- Paul: Yes. Might be better to talk to Apple to add API.
- Paul: One thing we can do is capture the output of the
browser. We use that for unit testing.
- [Paul to summarize in the issue]
- AnalyserNode: provide access to complex FFT result
<https://github.com/WebAudio/web-audio-api-v2/issues/107>
- Paul: Same problem as before. You don't know what you're
getting and when.
- Raymond: Yeah, you really can't use this for analysis.
- Hongchan: What's the best web impl of FFT?
- Paul: asm.js based from long ago. But PFFFT compiled to WASM
w and w/o SIMD is probably good.
- Hongchan: Maybe that would be best if we could point to a
good web FFT impl.
- Raymond: What's the conclusion? Won't do this because it
probably won't do what they really want. And recommend
using WASM with
PFFFT.
- Hongchan: That's a big ask!
- Raymond: Yeah, but that's not WebAudio's problem. :-)
- [Raymond to summarize]
- AnalyserNode: efficiency improvements to FFT post-processing
<https://github.com/WebAudio/web-audio-api-v2/issues/108>
- Hongchan: The node was originally for eye-candy
- Paul: I wonder if Meyda does analysis, maybe with worklets.
- Raymond: Conclusion? Close it as being too late to change the
processing.
- [Raymond to close with comment]
- Issues
- NoiseGenerator
<https://github.com/WebAudio/web-audio-api-v2/issues/8>: Updates
on using bigint
- Paul: No progress
- Raymond: Could take forever for browsers to support this
fully and not just in JS.
- 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
- Philippe: From what I understand, there's a proposed API.
- Philippe: Why was setSinkID added?
- Paul: I don't really remember, but it can be useful if you
want to switch but don't care about the impact on sample
rate and latency.
- Hongchan: Context option is good for getting optimum values.
setSinkID will be like a media fallback with resampling
and everything.
- Philippe: Relies on Media output
- Paul: Media Capture which allows enumerating devices.
- Philippe: Don't see need for setSinkID
- Hongchan: What happens if you plug in a device?
- Philippe: I expect the system to inform me
- Philippe: Allows setting a virtual output device for main
output? Say headphones are plugged in and I wouldn't want
to get a dialog
to select something.
- Hongchan: That can be done programatically.
- Paul: Changing default audio already handled
- Hongchan: If important for devs, context options is useful.
And setSinkID can be beneficial
- Paul: We have an internal impl of this that works.
- Phillippe: Wondering about device connection/disconnection
- Raymond: That's all part of the design of the API.
- Paul/Hongchan: Yeah, we need to get all of this done
- Philippe: If device selection UI pops up and user chooses usb
headphones and gets ID. Creates context with ID. Then
user unplugs usb
device. What happens? Fallback to default device? Popup
dialog to let
user choose a new device?
- Paul: On some native apps, engine stops. But reaper will
switch automatically to default.
- Hongchan: I'm wondering if, say, channel count changes.
- Paul: Yeah, we have no way to notify user about this.
- Raymond: I kind of like it if I'm listening to music in
headphones and I unplug it. audio stops.
- Paul: Yeah.
- Hongchan: Do you have the docs?
- Philippet: Here: ttps://
developer.apple.com/documentation/avfoundation/avaudioengine
- Paul: I agree with you Ray on stopping. It's really annoying
if it doesn't.
- Hongchan: But games will probably want to swtich seamlessly.
- Philippe: Yeah, the behavior depends on the app.
- [Time is up]
- AOB
- Decided not to have a meeting next week. The next meeting will be
Jan 2021, date TBD.
See you all next year!
On Wed, Dec 9, 2020 at 4:13 PM Raymond Toy <rtoy@google.com> wrote:
> Here's the agenda for our WG/CG meeting.
>
> Meeting venue https://meet.google.com/wgg-wjtd-kdu
>
> Agenda:
>
> - V1
> - CR Status
> - 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 #2032: ABSN extrapolates the last output
> <https://github.com/WebAudio/web-audio-api/pull/2256>
> - V2
> - New issues
> - AudioOutputContext
> <https://github.com/WebAudio/web-audio-api-v2/issues/106>
> - AnalyserNode: provide access to complex FFT result
> <https://github.com/WebAudio/web-audio-api-v2/issues/107>
> - AnalyserNode: efficiency improvements to FFT post-processing
> <https://github.com/WebAudio/web-audio-api-v2/issues/108>
> - 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, 10 December 2020 18:08:36 UTC