[Help] Use raw microphone input (Unprocessed / Media usage)

[Mr-Precise]

Hello ALVR team! Thank you for the amazing work you’ve done with the project — it’s truly fantastic ❤️

I’m currently trying to bring back a feature that used to work in earlier versions of the client: capturing raw microphone input from the Quest 2, without Android's automatic processing (noise suppression, echo cancellation, AGC, etc). I am using Linux *ubuntu 22.04 based OS.

In the past, this could easily be achieved in alvr/client_core/src/audio.rs by adding these two lines to the audio input stream builder:

.set_usage(Usage::Media)
.set_input_preset(InputPreset::Unprocessed)

That combination worked perfectly: the microphone signal was clean, unaltered, and loud enough — very useful for use cases like streaming, voice analysis, or recording background sounds. Additional: Would it be possible to expose a simple software gain multiplier?

Unfortunately, the current nightly builds have a fully rewritten audio.rs, and I’m struggling to figure out where or how to apply this behavior again. I’m not very familiar with Rust, so I'm having trouble adapting the changes myself.

Would it be possible to restore this functionality in the new codebase? Or at least point me in the right direction for implementing it again?

Thanks so much in advance!

[FaterSteve]

A gain multiplier in settings would be huge for me, and would all but resolve the bug report I posted earlier today github.com/alvr-org/ALVR/issues/3001