There's no rollback procedure in case a device switch fails right now,
nor does the code entrypoints that call the switching procedures wait
for resolution or failure before marking the new device as chosen. That
may cause inconsistent states in a couple of ways:
- No rollback: switch fails, audio is still on but no actual
microphone input is being transmitted
- Not waiting for resolutions: inconsistent chosen devices on failures
Device switching errors are also not surfaced to the end user
This commit:
- Adds device rollback and proper resolution/failure response
awaits to try and make the state a bit more consistent.
- Centralizes the input device switching code to be reused between
different bridges
- Centralizes device ID state management in audio-manager to try and
mantain them a bit more consistent across the board
- Surface device switching failures to the end user
- Guarantee device IDs are set to the session storage on all
appropriate scenarios
For scenarios where streams are produced in AudioSettings (local echo,
volume meter), force gUM resolution before devices are enumerated.
This effectively guarantees that all devices are present, labelled and
with deviceIds.
public.media.showVolumeMeterInSettings => public.media.showVolumeMeter
public.media.simplifiedEchoTest => public.media.localEchoTest.enabled
Initial hearing state can be configured in public.media.localEchoTest.initialHearingState
New features:
- A simplified echo test mode that only does a local loopback (instead of
going to FS and back)
- A volume meter for microphone streams to the AudioSettings view
Those two features are experimental and disabled by default; see
public.app.media.simplifiedEchoTest and public.app.media.showVolumeMeter configs
Collateral changes:
- fix: localize fallback device strings in AudioSettings/DeviceSelector
- Refactor on some media stream utils to be re-usable across components
- Refactor in AudioSettings to keep gUM #uses stable.
* TODO: need to pass streams through AudioManager to avoid the surplus gUM.
- fix(audio): drop ScriptProcessorNode usage (deprecated)
* Used in volume meter for tracking - use hark instead
We are now leaving the check for the minBrowserVersions object in settings.yml
If the settings enables chrome iOS, audio should allow users to be joining
with audio.
This is related to recent Chrome update (iOS 14.3+) that now allows
camera/microphone to be captured. We are looking for enabling this for
Chrome 93 in iOS (chromeMobileIOS version in settings.yml)
When listenOnlyMode=false, skipCheck=true and skipCheckOnJoin=true, the
audio tries to start a session more than one time, causing it to fail
at the first one (and reconnect after that).
Now we check if user is already connecting before trying to start a new
audio session.
Added some info in settings.yml for the options related to this commit
Closes#12190
When setting skipCheckOnJoin to true, an extra audio channel is created in
FreeSWITCH, after user accepts the echo test. The extra channel is removed
when user leaves the room, but this still may affect performance.
When joining/returning breakouts, audio would always connect
with full audio. This can lead to a performance problem, once
all listenonly users would join full audio, increasing the
number of streams in FreeSWITCH.
We now have a consistent behavior, which is:
1 - The choice made by the user in the main room is predominant:
if mic is active in main room, user will automatically
join mic in breakout room. When returning from breakout
room, user will also join with mic again.
2 - Changes made in breakout room won't have effect when
returning to the main room. This means if user, for example,
change from listenonly to mic in breakout room, the returning
will consider the option choosen previously (listenonly) and
listenonly will be active again in the main room.
3 - If user didn't join audio in the main room, the audio modal
will be prompted when joining the breakout room (this is
a special case of (1))
The following is some technicall information:
InputStreamLiveSelector (component.jsx) now calls
'handleLeaveAudio' function, which is the default
function when user leaves audio (also used when
dynamic devices are inactive).
We now store information about user's choice (mic or listenonly)
using local storage, instead of the previous cookie method (this
was triggering some warnings in browser's console).
Also did a small refactoring to match eslint rules.
Fixes#11662.
