I"m working on a project involving Tauri for the UI and webrtc-rs. A bird-eye overview of the steps to get it working are as follows:

1. You should get started by writing a WebRTC client in the Rust layer. See this for a more complete example: https://github.com/webrtc-rs/examples/pull/15/files
2. Create a struct to manage your call state and initilize it in your Tauri app"s src/main.rs, then give it to app.manage for Tauri to make it available for you to tauri command functions as a param so you can call methods on it. (example)
3. Utilize Tauri commands (invoke) from your front-end to manage your call state (e.g. to get media, create peer, remove peer, end call)
4. Get audio using cpal crate (best audio lib). ( Optionally you can use https://github.com/tonarino/webrtc-audio-processing to do echo cancallation)
5. Get camera video using something like nokhwa (actively developed)
6. Get screen capture from the scrap subcrate in this project which is an updated fork: https://github.com/rustdesk/rustdesk/tree/master/libs/scrap
7. Encode/decode audio using opus and video using libvp9 or another encoder. I haven"t got to this part yet.
8. Render decoded video in an egui window in Tauri: https://github.com/tauri-apps/tauri-egui (because you can"t access video frames from the webview, you need to use another kind of window rendering engine called egui which is managed in Rust. It"s not that hard to get working.

Hope it helps.