[REQUEST]: Configurable audio options (noise suppression, auto gain, etc.)

[rollersteaam]

Is your feature request related to a problem?

There is a ton of white noise and the user gets too loud when the phone's volume is above 50% when they join a call. This is for the scenario where you're transmitting the user's microphone. This can make for some very unpleasant calls.

Describe the solution you'd like

Let us access this in the nice C# land interface:

I have set these to true myself and built for Android and tested, but nothing happened.

I then tried to supply an Audio Processing module to the peer connection factory, but this also changed nothing:

It seems the solution may need to change the OnData implementation in UnityAudioTrackSource to use the APM.

Describe alternatives you've considered

Detecting the user's current volume and showing a warning when it becomes too high that it will cause a poor call experience. This isn't a good solution though.

Additional context

No response

[karasusan]

@rollersteaam
Sure, the noise cancelling is a essential feature for audio streaming. I want we will support it in the future.

[karasusan]

memo: WRS-152

[burakkaraceylan]

Hello! Any updates on this?

[kannan-xiao4]

@burakkaraceylan
No update, We give high priority to improving stability at present.

[wyliefoxxx]

This is an unfortunate dead-end to find... after spending the last week building this library into our project for use as a group chat system.. it looks like I/we are just going to have to wait until the echo cancellation feature is available to use it?? The system performs pretty horrendously as a chat system at the moment, as endless echoing pretty much ruins its usability. I appreciate the work the Unity team is doing, and has done on the system, but now I'm in hot water as the developer on the team who pushed us to go down this path. Until there's a fix for this serious issue, I hope you'll add some kind of warning to the documentation about it..

[rollersteaam]

@wyliefoxxx @kannan-xiao4 @karasusan

This is an unfortunate dead-end to find... after spending the last week building this library into our project for use as a group chat system.. it looks like I/we are just going to have to wait until the echo cancellation feature is available to use it?? The system performs pretty horrendously as a chat system at the moment, as endless echoing pretty much ruins its usability. I appreciate the work the Unity team is doing, and has done on the system, but now I'm in hot water as the developer on the team who pushed us to go down this path. Until there's a fix for this serious issue, I hope you'll add some kind of warning to the documentation about it..

A warning should be added, agreed. I am very thankful for the work these guys have done so far. However, I had previously fell into the exact same professional hole, and I think XR startups are very vulnerable to this in general. I would not consider this library ready for production use for Android or iOS platforms.

I understand that you have to install this package as a "prerelease" (thus, not production ready) but it is too open to interpretation. A warning in the README should say "Not ready for production use" or similar, perhaps detailing why (no noise cancellation) as essential capabilities of a WebRTC library you would see anywhere else are not supported here.

For unrelated reasons I have left my job which involved use of this library, so I will not be able to fix this issue stably myself. If I get time I may try to solve this challenge independently as a mark of pride :P But don't count on it.

[stay-cnsj]

Our team is having the same issue and this feature is on hold for our product. I really hope that echo cancellation will be implemented in the next version.